<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	>

<channel>
	<title>Technology Archives - Abstra</title>
	<atom:link href="https://abstra.co/tag/technology/feed/" rel="self" type="application/rss+xml" />
	<link>https://abstra.co/tag/technology/</link>
	<description>Your Nearshore Tech Partner</description>
	<lastBuildDate>Wed, 11 Nov 2026 19:10:01 +0000</lastBuildDate>
	<language>en-US</language>
	<sy:updatePeriod>
	hourly	</sy:updatePeriod>
	<sy:updateFrequency>
	1	</sy:updateFrequency>
	<generator>https://wordpress.org/?v=7.1.4</generator>

<image>
	<url>https://abstra.co/wp-content/uploads/2024/08/cropped-favicon-32x32.jpg</url>
	<title>Technology Archives - Abstra</title>
	<link>https://abstra.co/tag/technology/</link>
	<width>32</width>
	<height>32</height>
</image> 
	<item>
		<title>What Is an SDET?</title>
		<link>https://abstra.co/blog/blog-what-is-an-sdet/</link>
		
		<dc:creator><![CDATA[Abstra Team]]></dc:creator>
		<pubDate>Wed, 11 Nov 2026 19:09:55 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[AI]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482062104</guid>

					<description><![CDATA[<p>What is an SDET? A developer whose code tests your code, automatically, on every change. The same role also goes by names like SET, Developer in Test and QA Automation. As AI writes more code, the SDET decides how all of it gets checked.</p>
<p>The post <a href="https://abstra.co/blog/blog-what-is-an-sdet/">What Is an SDET?</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Ask &#8220;what is an SDET?&#8221; and most tech leaders will give you the short answer: a developer who writes automated tests. That answer is right, and it leaves out most of the story.</p>



<p class="wp-block-paragraph">The role has roots at Microsoft in the 1990s, goes by several job titles depending on the company, and today covers far more than writing tests: test architecture, release pipelines, code review and, more and more, how AI-generated code gets checked before it goes live. If you&#8217;re deciding whether your team needs one, or how to hire for it, this is the full picture.</p>



<h3 class="wp-block-heading">So, what is an SDET?</h3>



<p class="wp-block-paragraph">SDET stands for Software Development Engineer in Test: a developer whose main product is quality.</p>



<p class="wp-block-paragraph">In practice, an SDET writes code that tests your software automatically on every change, builds and maintains the testing framework, connects it to your release pipeline, and works next to your developers from the first line of code (<a target="_blank" rel="noopener" href="https://testomat.io/blog/what-is-sdet-software-development-engineer-in-test-explained/">Testomat.io</a>).</p>



<p class="wp-block-paragraph">The simplest way to frame it: your developers own the features, and the SDET owns the system that proves those features still work after every change.</p>



<h3 class="wp-block-heading">One role, many name tags</h3>



<p class="wp-block-paragraph">Even leaders who know the role well don&#8217;t always recognize it in a job post. Every company names it a little differently, and the titles overlap from team to team (<a target="_blank" rel="noopener" href="https://testomat.io/blog/what-is-sdet-software-development-engineer-in-test-explained/">Testomat.io</a>). These are the versions you&#8217;ll see most:</p>



<ul class="wp-block-list">
<li><strong>SDET (Software Development Engineer in Test):</strong> the original title. Microsoft pioneered it, Microsoft and Amazon made it popular, and today it shows up in job openings around the world (<a href="https://newsletter.pragmaticengineer.com/p/how-microsoft-does-quality-assurance" target="_blank" rel="noopener">The Pragmatic Engineer</a>; <a href="https://dev.to/automationhacks/who-the-heck-is-an-sdet-1hf4" target="_blank" rel="noopener">DEV Community</a>).</li>



<li><strong>SET (Software Engineer in Test):</strong> Google&#8217;s version. Same idea, shorter name: Google dropped the &#8220;D,&#8221; and many companies followed (<a href="https://dev.to/automationhacks/who-the-heck-is-an-sdet-1hf4" target="_blank" rel="noopener">DEV Community</a>).</li>



<li><strong>Developer in Test (DiT):</strong> common in teams where testing lives inside the dev team. Same daily work, with a title that leans developer (<a href="https://testomat.io/blog/what-is-sdet-software-development-engineer-in-test-explained/" target="_blank" rel="noopener">Testomat.io</a>).</li>



<li><strong>Test Automation / QA Automation:</strong> all over job posts. Often writes automated tests inside a framework someone else built (<a href="https://testomat.io/blog/what-is-sdet-software-development-engineer-in-test-explained/" target="_blank" rel="noopener">Testomat.io</a>).</li>



<li><strong>Quality Engineer, QA Engineer or Test Engineer:</strong> other titles companies use for similar work when they don&#8217;t have an SDET title (<a href="https://dev.to/automationhacks/who-the-heck-is-an-sdet-1hf4" target="_blank" rel="noopener">DEV Community</a>).</li>



<li><strong>Engineering Productivity:</strong> the name of the Google organization where testing lives, built around making other developers more productive and more quality-minded (<a href="https://dev.to/mszeles/how-google-tests-software-what-we-can-learn-from-google-a-book-extract-by-miki-szeles-4848" target="_blank" rel="noopener">How Google Tests Software, via DEV Community</a>).</li>
</ul>



<p class="wp-block-paragraph">So when you hire, skip the title debate and ask one question: does this person build the testing system, or only write tests inside one? A senior SDET does both, and that&#8217;s the difference you&#8217;ll feel by the second sprint.</p>



<h3 class="wp-block-heading">The SDET origin story, in four scenes</h3>



<ol class="wp-block-list">
<li><strong>Scene 1, 1979: a book flips the question.</strong> Glenford Myers publishes <em>The Art of Software Testing</em> with a bold idea for its time: you test software to find errors on purpose, not to prove it works. In his words, testing is &#8220;executing a program with the intent of finding errors&#8221; (<a href="https://agileconnection.com/print/book/art-software-testing" target="_blank" rel="noopener">AgileConnection</a>). Every SDET still works with that mindset today.</li>



<li><strong>Scene 2, the 1990s to 2004: Microsoft builds the role.</strong> Microsoft pioneered the SDET, most likely in the 1990s: a developer who wrote test code instead of product code, working on the same team as the developers. By 2004, its Exchange team was publicly explaining what the job meant (<a href="https://newsletter.pragmaticengineer.com/p/how-microsoft-does-quality-assurance" target="_blank" rel="noopener">The Pragmatic Engineer</a>). Google built its own version, the SET. And here&#8217;s a fun one: when Alberto Savoia joined Google as an engineering director in 2001, the company had about two hundred developers and three testers, according to his foreword to <em>How Google Tests Software</em> (<a href="https://dev.to/mszeles/how-google-tests-software-what-we-can-learn-from-google-a-book-extract-by-miki-szeles-4848" target="_blank" rel="noopener">DEV Community</a>). Three.</li>



<li><strong>Scene 3, 2004: Selenium and a chemistry joke.</strong> At ThoughtWorks in Chicago, Jason Huggins built a tool called &#8220;JavaScriptTestRunner&#8221; to test an internal time-and-expenses app (<a href="https://selenium.dev/about/" target="_blank" rel="noopener">Selenium</a>). Then a colleague joked that the new tool could &#8220;kill Mercury,&#8221; a rival testing product. Huggins went hunting for puns, found that selenium supplements are a cure for mercury poisoning, and replied: maybe we should name it Selenium. Everyone started calling it that right away. The best part? The rest of his email argued for a different name, &#8220;CheckEngine,&#8221; and everyone ignored it, as Huggins himself told the <a href="https://groups.google.com/g/selenium-users/c/17yVn5h7HzI" target="_blank" rel="noopener">Selenium Users group</a>. More than twenty years later, Selenium still holds roughly 22% of the QA automation market (<a href="https://testomat.io/blog/what-is-sdet-software-development-engineer-in-test-explained/" target="_blank" rel="noopener">Testomat.io</a>).</li>



<li><strong>Scene 4, 2026: AI writes the code, and someone still checks it.</strong> In a January 2026 SmartBear survey of 273 testing and quality decision-makers, 93% said their teams already use AI coding tools. At the same time, 68% were concerned that faster AI development will create testing bottlenecks (<a href="https://smartbear.com/ai-software-quality-gap-report/" target="_blank" rel="noopener">SmartBear</a>). In other words, more code arrives faster than ever, and the person who designs how it all gets checked has never mattered more.</li>
</ol>



<h3 class="wp-block-heading">A week in the life of an SDET</h3>



<p class="wp-block-paragraph">At a senior level, an SDET rarely runs tests by hand. Most of their week goes into designing how testing works and reviewing the team&#8217;s code (<a target="_blank" rel="noopener" href="https://testomat.io/blog/what-is-sdet-software-development-engineer-in-test-explained/">Testomat.io</a>). Here&#8217;s one way that week could look:</p>



<ul class="wp-block-list">
<li><strong>Monday: builds the testing system.</strong> Picks the tools, sets the patterns, and keeps everything current as the product grows.</li>



<li><strong>Tuesday: writes automated tests at every level.</strong> From tiny checks on one function to full user journeys, like signing up or paying.</li>



<li><strong>Wednesday: wires testing into your release pipeline.</strong> As a result, every change gets checked on its own, and a failing check blocks the change from merging.</li>



<li><strong>Thursday: fixes the flaky tests.</strong> Tests that pass one day and fail the next get repaired or retired, so the team trusts every green check.</li>



<li><strong>Friday: reviews pull requests and reports on quality.</strong> The SDET spots code that will be hard to test, then shares pass rates and coverage gaps in a view the whole team can act on.</li>
</ul>



<p class="wp-block-paragraph">Of course, real weeks mix all of this together. The result stays the same, though: a team that releases often and knows exactly what&#8217;s going out the door.</p>



<h3 class="wp-block-heading">Signs your team is ready for an SDET</h3>



<p class="wp-block-paragraph">If a few of these sound familiar, an SDET is probably the missing piece:</p>



<ul class="wp-block-list">
<li>You release an update, and the next day goes to fixing whatever it changed.</li>



<li>Your customers are the ones telling you something isn&#8217;t working.</li>



<li>Your developers spend more time on old bugs than on new features.</li>



<li>Testing everything by hand takes days, so launch dates keep moving.</li>



<li>Your team writes code with AI, and you want to be sure it all works before it goes live.</li>



<li>There&#8217;s a part of your system the team prefers not to touch, because the last change there caused trouble.</li>
</ul>



<p class="wp-block-paragraph">Good news: none of these say anything about your developers&#8217; talent. They simply show a team that has grown faster than the way it checks its work, and closing that gap is exactly what an SDET does best.</p>



<h3 class="wp-block-heading">Why AI makes the SDET more valuable</h3>



<p class="wp-block-paragraph">AI agents can now draft tests from code changes, decide which tests to run first and explore an app looking for bugs. Great news, right? It is. However, someone still has to decide what&#8217;s worth testing, what &#8220;risky&#8221; means for your product, and which AI-generated tests deserve your trust (<a target="_blank" rel="noopener" href="https://testomat.io/blog/what-is-sdet-software-development-engineer-in-test-explained/">Testomat.io</a>). That someone is the SDET.</p>



<p class="wp-block-paragraph">Tools alone haven&#8217;t closed the gap either. In the same SmartBear survey, 87% of teams had automated at least a fifth of their testing, and 92% still tested manually (<a target="_blank" rel="noopener" href="https://smartbear.com/ai-software-quality-gap-report/">SmartBear</a>). The tools are already on the shelf. What many teams are missing is the person who designs how they all work together, the same senior judgment we wrote about in&nbsp;<a target="_blank" rel="noopener" href="https://abstra.co/blog/ai-coding-agent-oversight/">AI coding agent oversight</a>&nbsp;and&nbsp;<a target="_blank" rel="noopener" href="https://abstra.co/nearshore-software-development-ai-era/">the part of nearshore that stays human</a>.</p>



<p class="wp-block-paragraph">Our take: as AI writes more of the code, quality becomes a design decision, and a senior SDET is the professional who owns it.</p>



<h3 class="wp-block-heading">Senior SDETs from Latin America, with Abstra</h3>



<p class="wp-block-paragraph">So, what is an SDET, in one line? The professional who lets your team build fast and trust every release.</p>



<p class="wp-block-paragraph">At Abstra, we work with top-tier senior SDETs across Latin America, with exactly the profile this article describes. They join your team working your hours and fully dedicated to your product.</p>



<p class="wp-block-paragraph">Thinking about adding one? Send us a message, and we&#8217;ll help you find the right person.</p>



<p class="wp-block-paragraph">That&#8217;s Abstra.</p>



<h3 class="wp-block-heading">FAQ</h3>



<p class="wp-block-paragraph"><strong>What is an SDET?</strong>&nbsp;An SDET is a developer who writes code that tests software automatically, builds the testing system around it, and connects it to the team&#8217;s release pipeline.</p>



<p class="wp-block-paragraph"><strong>What does SDET stand for?</strong>&nbsp;Software Development Engineer in Test: a developer who writes code to test software automatically.</p>



<p class="wp-block-paragraph"><strong>What&#8217;s the difference between an SDET and a QA tester?</strong>&nbsp;A QA tester checks the application, often by hand. An SDET writes the code and builds the system that checks it automatically, every time something changes.</p>



<p class="wp-block-paragraph"><strong>Is an SDET the same as a test automation engineer?</strong>&nbsp;They overlap. A test automation engineer usually writes tests inside an existing framework; an SDET designs the framework and decides what gets tested where.</p>



<p class="wp-block-paragraph"><strong>Does AI replace the SDET?</strong>&nbsp;No. AI speeds up writing tests, and the SDET decides what to test, defines what risk means for the product, and reviews what AI produces.</p>
<p>The post <a href="https://abstra.co/blog/blog-what-is-an-sdet/">What Is an SDET?</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Where You Shouldn&#8217;t Trust AI Without a Second Look</title>
		<link>https://abstra.co/blog/blog-ai-hallucinations-fintech/</link>
		
		<dc:creator><![CDATA[Abstra Team]]></dc:creator>
		<pubDate>Fri, 30 Oct 2026 17:44:58 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Business]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482062097</guid>

					<description><![CDATA[<p>AI hallucinations in fintech look like confident answers: an invented fee, a wrong reason for a loan denial, a figure or rule nobody wrote.</p>
<p>The post <a href="https://abstra.co/blog/blog-ai-hallucinations-fintech/">Where You Shouldn&#8217;t Trust AI Without a Second Look</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">AI is making fintech teams faster every week. It answers customers at 2 a.m., drafts reports in minutes and helps explain credit decisions at scale. That speed is real, and it&#8217;s worth keeping.</p>



<p class="wp-block-paragraph">However, AI hallucinations in fintech rarely look like mistakes. They look like answers: a fee quoted with total confidence, a loan denial explained in clear language, a compliance summary that cites a rule by name. The problem is that the fee, the reason or the rule may not exist.</p>



<p class="wp-block-paragraph">So the useful question is where an answer needs a second look before it reaches a customer, a regulator or a board. Here are the three places that matter most, and how teams catch hallucinations there without slowing down.</p>



<h2 class="wp-block-heading">What an AI hallucination is, and why it happens</h2>



<p class="wp-block-paragraph">An AI hallucination is an answer that sounds right but isn&#8217;t. The model invents a policy, a source or a number and presents it with the same confidence as a fact.</p>



<p class="wp-block-paragraph">This happens because language models predict the most likely next words. Most of the time, likely and true line up. Sometimes they don&#8217;t, and the output is simply convincing. In most industries that&#8217;s an annoyance. In fintech, the answer often touches someone&#8217;s money, credit or legal rights, so convincing isn&#8217;t enough.</p>



<h2 class="wp-block-heading">1. Customer support: when a chatbot makes a promise</h2>



<p class="wp-block-paragraph">A support chatbot can invent a fee, an interest rate or a refund policy that sounds completely official. Your customer has no way to tell the difference. They read it as a promise from your company.</p>



<p class="wp-block-paragraph">This isn&#8217;t hypothetical. In 2024, a British Columbia tribunal ruled against Air Canada after its website chatbot described a bereavement refund policy that didn&#8217;t match the airline&#8217;s real one. The airline argued it shouldn&#8217;t be responsible for what the chatbot said. The tribunal disagreed, found the airline hadn&#8217;t taken reasonable care to keep the chatbot accurate, and ordered it to pay the difference (<a target="_blank" rel="noopener" href="https://www.cbc.ca/amp/1.7116416">CBC News</a>;&nbsp;<a target="_blank" rel="noopener" href="https://www.law360.ca/ca/articles/1804075">Law360 Canada</a>).</p>



<p class="wp-block-paragraph">It was an airline, not a bank. Still, the lesson carries straight into fintech: customers treat a chatbot&#8217;s answer as the company&#8217;s answer. When the topic is a fee, a rate or a dispute window, that answer should come from your real policy, not from the model&#8217;s best guess.</p>



<h2 class="wp-block-heading">2. Credit decisions: when the explanation sounds right but isn&#8217;t the real reason</h2>



<p class="wp-block-paragraph">When a lender uses AI to help score applications or draft the letters that explain a decision, a quiet kind of hallucination can show up: a denial explained with a reason that reads well but isn&#8217;t what drove the decision.</p>



<p class="wp-block-paragraph">In the U.S., that gap matters. Under Regulation B, which implements the Equal Credit Opportunity Act, a lender that denies credit must give specific principal reasons (<a target="_blank" rel="noopener" href="https://www.ecfr.gov/current/title-12/chapter-X/part-1002/subpart-A/section-1002.9">12 CFR 1002.9, eCFR</a>). The official interpretation adds that those reasons must accurately describe the factors the creditor considered or scored (<a target="_blank" rel="noopener" href="https://www.consumerfinance.gov/rules-policy/regulations/1002/interp-9/">CFPB, Regulation B comment 9(b)(2)-2</a>). The rule makes no exception for decisions an AI model helped make.</p>



<p class="wp-block-paragraph">In practice, a generated explanation is a draft. Before it goes out, someone has to confirm that each reason maps to a real factor in the model, not to the most plausible sentence the AI could write.</p>



<h2 class="wp-block-heading">3. Reports and compliance: figures nobody calculated, rules nobody wrote</h2>



<p class="wp-block-paragraph">AI is a great first-draft tool for financial summaries and compliance research. It&#8217;s also where invented numbers and invented sources are hardest to spot, because they arrive inside polished, well-formatted text.</p>



<p class="wp-block-paragraph">Two public examples show the pattern. In 2023, researchers at Patronus AI tested leading models on clear-cut questions about real SEC filings. Even the best setup, with almost the whole filing in front of it, answered about 79% of questions correctly and gave wrong answers to roughly 17% (<a target="_blank" rel="noopener" href="https://dc.fortune.com/2023/12/21/chatgpt-understand-sec-filings-anthropic-meta-llama2-openai-finance-ai">Fortune</a>;&nbsp;<a target="_blank" rel="noopener" href="https://arxiv.org/abs/2311.11944">FinanceBench paper</a>). Models have changed a lot since 2023, but the lesson holds: a confident figure still needs a source.</p>



<p class="wp-block-paragraph">The same year, a federal judge in New York fined two lawyers $5,000 after they filed a brief citing court decisions that ChatGPT had made up (<a target="_blank" rel="noopener" href="https://www.seyfarth.com/news-insights/update-on-the-chatgpt-case-counsel-who-submitted-fake-cases-are-sanctioned.html">Seyfarth Shaw</a>). A compliance memo that cites a regulation that doesn&#8217;t exist works the same way. It reads perfectly until someone goes to look it up.</p>



<h2 class="wp-block-heading">How to catch AI hallucinations before they reach a customer</h2>



<p class="wp-block-paragraph">The good news is that the fix is a process, and most of it is familiar to any team that ships regulated software. Four habits cover the three areas above:</p>



<ol class="wp-block-list">
<li><strong>Ground answers in your own data.</strong> Connect the chatbot to your actual fee schedule, rate tables and policies, and have it say &#8220;I don&#8217;t know&#8221; when the answer isn&#8217;t there.</li>



<li><strong>Test the output before every release.</strong> Keep a set of real questions with known correct answers, especially about money, and run them every time the model, prompt or data changes.</li>



<li><strong>Trace every reason and every number.</strong> A credit explanation should map to a real model factor. A figure in a report should link to the cell or filing it came from. A cited rule should link to the rule.</li>



<li><strong>Put a person on anything that touches money.</strong> Let AI draft at full speed, then have someone who knows the domain approve what goes to a customer, a regulator or a board.</li>
</ol>



<p class="wp-block-paragraph">None of this slows a team down much. It moves the checking to the moments where a wrong answer would cost the most.</p>



<h2 class="wp-block-heading">How Abstra approaches it</h2>



<p class="wp-block-paragraph">At Abstra, we see AI as the best accelerator a product team has had in years. It writes the boilerplate fast. What stays human is the judgment: system design, code review and catching what a confident AI gets wrong (<a target="_blank" rel="noopener" href="https://abstra.co/nearshore-software-development-ai-era/">Nearshore in the AI Era: The Part That Stays Human</a>).</p>



<p class="wp-block-paragraph">That&#8217;s why we place senior tech talent from Latin America on our clients&#8217; teams, working the same hours they do. In fintech, that seniority is what turns AI speed into answers you can stand behind, because every answer that touches money gets a second look from someone who knows what to check.</p>



<p class="wp-block-paragraph">If your team is adding AI to support, lending or reporting and wants that second look built in from day one, let&#8217;s talk.</p>



<h2 class="wp-block-heading">FAQ</h2>



<p class="wp-block-paragraph"><strong>What are AI hallucinations in fintech?</strong>&nbsp;They are AI answers that sound correct but aren&#8217;t, in places where money is involved: an invented fee in a chatbot reply, a credit denial explained with the wrong reason, or a report with figures or rules that don&#8217;t exist.</p>



<p class="wp-block-paragraph"><strong>Where do AI hallucinations cause the most risk for a fintech?</strong>&nbsp;Customer support, credit decisions, and financial reporting or compliance, because customers and regulators treat those answers as the company&#8217;s official position.</p>



<p class="wp-block-paragraph"><strong>Do U.S. lending rules still apply when AI helps make a credit decision?</strong>&nbsp;Yes. Regulation B requires specific principal reasons for a denial that accurately describe the factors the creditor considered, whether or not an AI model was involved.</p>



<p class="wp-block-paragraph"><strong>How can a fintech reduce AI hallucinations?</strong>&nbsp;Ground the AI in your own data, test outputs before every release, trace every reason and number to its source, and have an experienced person approve anything that touches money.</p>
<p>The post <a href="https://abstra.co/blog/blog-ai-hallucinations-fintech/">Where You Shouldn&#8217;t Trust AI Without a Second Look</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Don’t Chase the Tools. Build the Instinct.</title>
		<link>https://abstra.co/blog/dont-chase-the-tools-build-the-instinct/</link>
		
		<dc:creator><![CDATA[Paola Brizuela]]></dc:creator>
		<pubDate>Wed, 21 Oct 2026 19:16:17 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Cybersecurity]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482062089</guid>

					<description><![CDATA[<p>Security tools only catch what they were built to catch. Instinct catches the rest.</p>
<p>The post <a href="https://abstra.co/blog/dont-chase-the-tools-build-the-instinct/">Don’t Chase the Tools. Build the Instinct.</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph"><em>Security isn’t bought. It’s built, and owned.</em></p>



<p class="wp-block-paragraph">On the first day of a project, you get access. On the last day, it goes away. Everything in between is where security lives.</p>



<p class="wp-block-paragraph">We asked Paola Brizuela, who leads cybersecurity at Abstra, what that looks like day to day. Her first answer had nothing to do with software:</p>



<blockquote class="wp-block-quote is-layout-flow wp-block-quote-is-layout-flow">
<p class="wp-block-paragraph">“Security takes more than buying tools. It requires visibility, clear owners, verifiable controls, and business continuity.”</p>
</blockquote>



<h3 class="wp-block-heading">Why instinct?</h3>



<p class="wp-block-paragraph">Security tools are like a smoke detector. They’re great at beeping when there’s smoke, but they won’t tell you that you left the front door open.</p>



<p class="wp-block-paragraph">That’s the part only people catch. It’s the same instinct that makes you check the stove before leaving home, or double-check that you didn’t just hit “reply all.” A scanner can’t remember that a former teammate still has the keys to the repo, but you can.</p>



<p class="wp-block-paragraph">On top of that, the house keeps getting new doors. A new integration here, a new account there, a shiny AI agent on Monday morning. No checklist grows that fast, but a good habit does.</p>



<p class="wp-block-paragraph">And the best part? Nobody is born with this instinct. It’s like learning to drive: at first you check every mirror on purpose, and one day you just do it without thinking. So let’s practice.</p>



<h3 class="wp-block-heading">Spot the open door</h3>



<p class="wp-block-paragraph">Which of these leave a door open?</p>



<p class="wp-block-paragraph">A. A contractor wrapped up a project three months ago. Their repository access is still active.<br>B. An admin account opens with just a password.<br>C. A new SaaS tool gets connected to production on a Friday afternoon, with no review.<br>D. Only one person on the team knows how deployment works.</p>



<p class="wp-block-paragraph"><strong>Answer: all four.</strong>&nbsp;Paola calls these the basics, and they’re where many breaches start. The good news is that every one has a simple fix.</p>



<p class="wp-block-paragraph">That fix starts the day you join a project and lasts until the day you leave.</p>



<h3 class="wp-block-heading">The life of an access</h3>



<p class="wp-block-paragraph"><strong>Day one:</strong>&nbsp;only the access the work needs.<br><strong>On the project:</strong>&nbsp;development, staging and production stay apart.<br><strong>Role change:</strong>&nbsp;permissions get reviewed.<br><strong>Last day:</strong>&nbsp;access ends with the project, on time.</p>



<p class="wp-block-paragraph">Simple habits, but they make a real difference.</p>



<p class="wp-block-paragraph"><strong>BY THE NUMBERS</strong><br><strong>62%</strong>&nbsp;of breaches involved the human element.<br><em>Source: Verizon 2026 Data Breach Investigations Report</em></p>



<p class="wp-block-paragraph">That number works both ways. If people are part of the risk, people are also the best defense. Which brings us to you.</p>



<h3 class="wp-block-heading">Your 30-second check</h3>



<p class="wp-block-paragraph">On the project you’re working on right now, can you answer these?</p>



<ol class="wp-block-list">
<li>What are the critical assets?</li>



<li>Who has access to them?</li>



<li>Which vendors handle the data?</li>



<li>What would happen if an essential system stopped working?</li>
</ol>



<p class="wp-block-paragraph">If you answered all four, that’s the instinct this edition is about.</p>



<p class="wp-block-paragraph"><strong>LAST LINE</strong><br>What’s one access you’d clean up today if you checked?</p>



<p class="wp-block-paragraph">Read Paola’s full article:&nbsp;<a href="https://abstra.co/blog/secure-by-design-software-projects/">https://abstra.co/blog/secure-by-design-software-projects/</a></p>
<p>The post <a href="https://abstra.co/blog/dont-chase-the-tools-build-the-instinct/">Don’t Chase the Tools. Build the Instinct.</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Secure by Design: How We Protect Every Project </title>
		<link>https://abstra.co/blog/secure-by-design-software-projects/</link>
		
		<dc:creator><![CDATA[Paola Brizuela]]></dc:creator>
		<pubDate>Tue, 06 Oct 2026 18:48:43 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Career Path]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482062079</guid>

					<description><![CDATA[<p>Most breaches start with the basics. At Abstra, every project is secure by design from day one.</p>
<p>The post <a href="https://abstra.co/blog/secure-by-design-software-projects/">Secure by Design: How We Protect Every Project </a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Most security problems don&#8217;t start with a sophisticated attack. They start with something small: a former contractor who still has access to a repository, an admin account without MFA, a new SaaS tool someone connected to production on a Friday afternoon.&nbsp;</p>



<p class="wp-block-paragraph">None of these are exotic. That&#8217;s exactly why they&#8217;re dangerous, and why we treat every project as secure by design from day one.&nbsp;</p>



<h3 class="wp-block-heading"><strong>Security by design in every project</strong>&nbsp;</h3>



<p class="wp-block-paragraph">Abstra doesn&#8217;t sell cybersecurity as a standalone service. We build software and cloud solutions with senior tech talent from Latin America. However, every one of those projects involves someone&#8217;s code, data, and infrastructure.&nbsp;</p>



<p class="wp-block-paragraph">Even though we don&#8217;t offer cybersecurity as a separate service, security has to be built into these projects across the board.&nbsp;</p>



<p class="wp-block-paragraph">In practice, that means:&nbsp;</p>



<ul class="wp-block-list">
<li><strong>Access control.</strong> People get the access they need for the work in front of them, and nothing more. </li>
</ul>



<ul class="wp-block-list">
<li><strong>Data protection.</strong> Sensitive data is handled deliberately, not by habit. </li>
</ul>



<ul class="wp-block-list">
<li><strong>Secure environment management.</strong> Development, staging, and production stay separate, and changes move between them with intention. </li>
</ul>



<ul class="wp-block-list">
<li><strong>Permission reviews.</strong> Access is reviewed and removed when roles change or a project ends. </li>
</ul>



<ul class="wp-block-list">
<li><strong>Good practices during development and operation.</strong> Security doesn&#8217;t stop at the pull request. It continues once the system is live. </li>
</ul>



<p class="wp-block-paragraph">Internally, we&#8217;re also strengthening our own information security, risk management, and compliance practices. We hold ourselves to the same standard we&#8217;d expect from anyone we hand our own data to.&nbsp;</p>



<h3 class="wp-block-heading"><strong>The mistakes I see most often</strong>&nbsp;</h3>



<p class="wp-block-paragraph">One of the most common mistakes I see is treating security as only a technical problem. Many breaches start with the basics.&nbsp;</p>



<p class="wp-block-paragraph">Those basics look like this:&nbsp;</p>



<ul class="wp-block-list">
<li>Access that isn&#8217;t revoked on time </li>
</ul>



<ul class="wp-block-list">
<li>Excessive permissions </li>
</ul>



<ul class="wp-block-list">
<li>No multi-factor authentication </li>
</ul>



<ul class="wp-block-list">
<li>No up-to-date inventory </li>
</ul>



<ul class="wp-block-list">
<li>Tools adopted without prior evaluation </li>
</ul>



<ul class="wp-block-list">
<li>Little training for the team </li>
</ul>



<ul class="wp-block-list">
<li>Processes that depend too much on what one person knows </li>
</ul>



<p class="wp-block-paragraph">There&#8217;s a second pattern, too, and it&#8217;s just as common.&nbsp;</p>



<p class="wp-block-paragraph">Companies often put controls in place only when a client, an audit, or an incident demands it. Security should be built into processes from the design stage and maintained continuously.&nbsp;</p>



<p class="wp-block-paragraph">When security waits for a trigger, it becomes a scramble.&nbsp;</p>



<h3 class="wp-block-heading"><strong>What I&#8217;m watching next: AI and SaaS</strong>&nbsp;</h3>



<p class="wp-block-paragraph">The attack surface is changing fast, and much of that change is coming from inside organizations.&nbsp;</p>



<p class="wp-block-paragraph">Every new integration, account, or AI agent can introduce access and risks that the organization isn&#8217;t managing yet.&nbsp;</p>



<p class="wp-block-paragraph">Because of this, it&#8217;s one of the areas I&#8217;m paying the closest attention to, both in our own operations and in the projects we support. The questions are simple, even if the answers take work: What does this tool have access to? Who approved it? What data does it see? What happens if it&#8217;s compromised?&nbsp;</p>



<h3 class="wp-block-heading"><strong>Four questions you should be able to answer today</strong>&nbsp;</h3>



<p class="wp-block-paragraph">If I had 30 seconds with a CTO, I&#8217;d tell them: don&#8217;t wait for an incident or a client questionnaire to start getting security in order.&nbsp;</p>



<p class="wp-block-paragraph">If you lead a tech team, you should be able to answer these right now:&nbsp;</p>



<ol start="1" class="wp-block-list">
<li>What are your critical assets? </li>
</ol>



<ol start="2" class="wp-block-list">
<li>Who has access to them? </li>
</ol>



<ol start="3" class="wp-block-list">
<li>Which vendors handle your data? </li>
</ol>



<ol start="4" class="wp-block-list">
<li>How would your company respond if an essential system stopped working? </li>
</ol>



<p class="wp-block-paragraph">Then look closely at your AI and SaaS tools, because that&#8217;s where new access appears fastest.&nbsp;</p>



<p class="wp-block-paragraph">Security takes more than buying tools. It requires visibility, clear owners, verifiable controls, and business continuity.&nbsp;</p>



<h3 class="wp-block-heading"><strong>What this means when you work with us</strong>&nbsp;</h3>



<p class="wp-block-paragraph">When you bring Abstra&#8217;s professionals into your team, you&#8217;re not handing your code and data to a black box. You&#8217;re working with people who treat access, permissions, and data as part of the work itself, and a company that keeps investing in getting those basics right.&nbsp;</p>



<p class="wp-block-paragraph">If security is what&#8217;s making you hesitate about nearshoring, let&#8217;s talk about it openly. It&#8217;s a fair concern, and it deserves a real conversation.&nbsp;</p>



<h3 class="wp-block-heading"><strong>FAQ</strong>&nbsp;</h3>



<ul class="wp-block-list">
<li><strong>Does Abstra offer cybersecurity services?</strong> Not as a standalone service. Security is built into the software development and cloud projects we deliver, through access controls, data protection, secure environment management, and permission reviews.&nbsp;</li>



<li><strong>What&#8217;s the most common security mistake companies make?</strong> Treating security as only a technical problem. Many breaches start with basics like unrevoked access, excessive permissions, or missing MFA.&nbsp;</li>



<li><strong>Why do AI tools increase security risk?</strong> Every new AI tool, integration, account, or agent can introduce access the organization isn&#8217;t tracking or managing yet.&nbsp;</li>
</ul>
<p>The post <a href="https://abstra.co/blog/secure-by-design-software-projects/">Secure by Design: How We Protect Every Project </a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Nearshore Without Moving Your Data</title>
		<link>https://abstra.co/blog/nearshore-data-residency/</link>
		
		<dc:creator><![CDATA[Abstra Team]]></dc:creator>
		<pubDate>Fri, 11 Sep 2026 17:53:33 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Business]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482062056</guid>

					<description><![CDATA[<p>Before an enterprise partner signs, they ask one thing: where does our data live, and who can see it from another country?</p>
<p>The post <a href="https://abstra.co/blog/nearshore-data-residency/">Nearshore Without Moving Your Data</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Say your company builds a health app or a payments app. You want to move faster, so you hire senior talent in Latin America. Good call. But then an enterprise partner, like a hospital or a bank, gets ready to sign. And they ask a question that decides everything: where does our data live, and who can see it from another country? That question is nearshore data residency.</p>



<p class="wp-block-paragraph">Here is the good news. Nearshore data residency sounds scary, but it is simple once you break it down. It is not really about where a person sits. It is about what you can control and prove.</p>



<h3 class="wp-block-heading">The question an enterprise partner asks first</h3>



<p class="wp-block-paragraph">Enterprise partners follow strict rules, and they pass those rules to you.&nbsp;<a href="https://www.hhs.gov/hipaa/index.html">HIPAA</a>&nbsp;covers health data.&nbsp;<a href="https://www.pcisecuritystandards.org/">PCI DSS</a>&nbsp;covers card data.&nbsp;<a href="https://gdpr.eu/">GDPR</a>&nbsp;covers personal data. So their security team checks you, and then checks anyone who works with you.</p>



<p class="wp-block-paragraph">They really only want three answers:</p>



<ul class="wp-block-list">
<li>Where does the data sit?</li>



<li>Who can reach it from another country?</li>



<li>Can you prove both?</li>
</ul>



<p class="wp-block-paragraph">Give clear answers, and the deal moves. Have them ready, and you stay a step ahead.</p>



<h3 class="wp-block-heading">Nearshore data residency is about control, not the country</h3>



<p class="wp-block-paragraph">Many people think &#8220;another country&#8221; means &#8220;against the rules.&#8221; It does not. The rules almost never ban access from abroad. They ask you to control where the data lives and who touches it.</p>



<p class="wp-block-paragraph">Think of it like lending your house. You do not hand over every key. Instead, one guest gets the key to a single room. A camera shows what they touched. And when they leave, you take the key back and you can prove it. Their nationality never mattered. Your control did.</p>



<p class="wp-block-paragraph">Your data works the same way. Two quick examples:</p>



<ul class="wp-block-list">
<li>A health startup sells to a hospital. The hospital asks for a signed <a href="https://www.hhs.gov/hipaa/for-professionals/privacy/guidance/business-associates/index.html">Business Associate Agreement</a> and proof that patient data never leaves its cloud. A good setup already covers both.</li>



<li>A fintech adds a bank partner. The bank asks where card data sits and who can open it. A logged, least-access setup answers on one screen.</li>
</ul>



<h3 class="wp-block-heading">What a good setup looks like</h3>



<p class="wp-block-paragraph">A partner who has done this before builds it to pass the check:</p>



<ul class="wp-block-list">
<li><strong>Data stays in your house.</strong> The team works inside your cloud and your region. Data never lands on someone&#8217;s personal laptop.</li>



<li><strong>They visit, they do not take it home.</strong> People reach the data through logged access, instead of copying it across a border.</li>



<li><strong>Locks on by default.</strong> You protect data in transit and at rest, so a remote connection is still a safe one.</li>



<li><strong>One key per room.</strong> Each person gets only what the task needs, and every action leaves a trail.</li>



<li><strong>Papers that match the promise.</strong> You get signed BAAs and DPAs, not just nice words.</li>



<li><strong>Clean goodbye.</strong> When someone leaves, access closes, and you can prove it.</li>
</ul>



<h3 class="wp-block-heading">How Abstra handles nearshore data residency</h3>



<p class="wp-block-paragraph">Abstra keeps one idea at the center. Your team should feel in-house and work inside your controls. That answers the residency question by design.</p>



<p class="wp-block-paragraph">Here is what Abstra brings:</p>



<ul class="wp-block-list">
<li><strong><a href="https://abstra.co/services-dedicated-teams/">Dedicated Teams</a> inside your systems.</strong> They work in your cloud, your logins, and your review steps, so data stays put. Need just one person? <a href="https://abstra.co/services-staff-augmentation/">Staff Augmentation</a> adds senior talent straight to your team.</li>



<li><strong>US-trained leaders</strong> who know how strict U.S. buyers think. Abstra brings 15+ years of nearshore experience, 4 U.S. locations, and 2 LATAM hubs.</li>



<li><strong>200+ vetted senior professionals across 15+ countries</strong>, with depth in <a href="https://abstra.co/our-services/solutions/ai-data/">Data and AI</a>, <a href="https://abstra.co/our-services/solutions/platform-infrastructure/">platform and infrastructure</a>, QA, and more.</li>



<li><strong>A simple 3-step start.</strong> Abstra runs a kickoff, plans your scope and access, then tracks the work. Most teams start in 5 to 7 business days, on U.S. time.</li>



<li><strong>Real work in regulated fields.</strong> U.S. finance, healthcare, and education teams already trust Abstra. Financial-services clients like Ascensus rely on Abstra to scale across their teams. <a href="https://abstra.co/our-clients/">See more clients</a>.</li>
</ul>



<p class="wp-block-paragraph">So the residency talk gets short. Abstra shows where data sits, who can reach it, and how access closes, before your buyer even asks.</p>



<h3 class="wp-block-heading">Answer it before they ask</h3>



<p class="wp-block-paragraph">Solve this early. When your partner explains, in plain words, where data sits and who can reach it, the security review shrinks and your buyer relaxes.</p>



<p class="wp-block-paragraph">A staffing pool often stops at the seat. That can leave someone working from a personal device, in an unknown place, with access nobody wrote down. Careful buyers look closely at that detail.</p>



<p class="wp-block-paragraph">Abstra closes it on purpose.&nbsp;<a href="https://abstra.co/contact/">Talk to Abstra</a>, or&nbsp;<a href="https://abstra.co/cost-calculator/">run the numbers</a>&nbsp;first.</p>
<p>The post <a href="https://abstra.co/blog/nearshore-data-residency/">Nearshore Without Moving Your Data</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>Secure Nearshore Software Development Starts on Day One</title>
		<link>https://abstra.co/blog/secure-nearshore-software-development-access/</link>
		
		<dc:creator><![CDATA[Abstra Team]]></dc:creator>
		<pubDate>Thu, 03 Sep 2026 13:38:33 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Business]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482062004</guid>

					<description><![CDATA[<p>Secure nearshore software development is a control problem, not a trust problem. Bound the access instead of granting blind trust.</p>
<p>The post <a href="https://abstra.co/blog/secure-nearshore-software-development-access/">Secure Nearshore Software Development Starts on Day One</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Secure nearshore software development does not fail at the interview. It fails at one quiet moment. Before your new tech professional joins a single standup, someone on your team grants a person they have never met access to the repository, the CI/CD pipeline, the secrets, and sometimes production or customer data.</p>



<p class="wp-block-paragraph">That decision is where nearshore plans quietly stall. Not at the interview. Not at the coding test. At the access grant.</p>



<p class="wp-block-paragraph">By now, you already believe the tech professional can build. The résumé is real, the code sample checked out, and the references were fine. Still, one question stays hard to answer, and your security team will eventually ask it. So will your customer&#8217;s security team. What exactly can this person see? What can they pull? Could you prove later precisely what an outside contributor touched?</p>



<h3 class="wp-block-heading">Skill was never the hard part</h3>



<p class="wp-block-paragraph">Most conversations about hiring remote or nearshore tech professionals focus on capability. Can they code? Do they communicate? Will the time zones work? In fact, those questions are solvable, and teams solve them every day.</p>



<p class="wp-block-paragraph">The harder part rarely gets said out loud, yet it is the one that actually holds up the requisition. The moment you grant access, a new blast radius opens inside your systems. As a result, for a person you do not yet know, that radius starts out invisible. What falls in scope? What got granted too broadly because that was faster than scoping it narrowly? If this person rolls off next month, what happens to the access? Can anyone confirm it was fully revoked?</p>



<p class="wp-block-paragraph">With an anonymous, rotating seat from a volume staffing vendor, you cannot answer any of that. The individual is interchangeable by design. They may sit on your project this quarter and on another company&#8217;s project next quarter. You inherited a name on an invoice, not an accountable person inside your controls. So the honest security answer to who touched this, and what they could reach, becomes &#8220;we are not entirely sure.&#8221; No serious CTO or CISO approves &#8220;we are not entirely sure.&#8221;</p>



<p class="wp-block-paragraph">So the requisition sits. The roadmap slips. And &#8220;we will handle the security piece later&#8221; becomes the thing that never happens.</p>



<h3 class="wp-block-heading">Secure nearshore software development is a control problem, not a trust one</h3>



<p class="wp-block-paragraph">Trust, stated in the abstract, is not decidable. You cannot verify a feeling before a customer audit. So stop trying to grant trust, and start bounding access instead. Turn a vague worry into a set of things you can see, scope, and prove.</p>



<p class="wp-block-paragraph">Framed that way, the access decision has real answers. Those answers depend on how you add the tech professional, not on how talented they are.</p>



<h3 class="wp-block-heading">What bounding access looks like in practice</h3>



<p class="wp-block-paragraph">Dedicated, named senior talent who works inside your own controls changes every part of the answer:</p>



<ul class="wp-block-list">
<li><strong>Your identity, not theirs.</strong> The tech professional signs in through your SSO and identity provider. They follow your policies, your MFA, and your session rules. Access becomes something you provision and watch, rather than a standing key you hand to a third party.</li>



<li><strong>Only the access the work needs.</strong> Access maps to the task, so it covers this repo, this pipeline, and this environment, and nothing wider. As a result, you avoid the broad grant that feels convenient at first and then gets forgotten by month three.</li>



<li><strong>Your process, your review gates.</strong> The tech professional commits into your branching model and your required reviews. Nothing reaches production on a side path.</li>



<li><strong>A footprint you can audit.</strong> The tech professional is a named person inside your systems. So you can answer what this contributor touched with a log, not a shrug. That is the exact evidence your security team and your customers ask for.</li>



<li><strong>Clean, provable removal of access.</strong> When the engagement changes, you revoke access, and you can show that you did. Therefore no key lingers, nothing stays ambiguous, and no seat quietly rotates toward someone else&#8217;s project.</li>
</ul>



<p class="wp-block-paragraph">None of that is exotic. In fact, you would expect the same posture from a good permanent hire. In short, this posture comes built in with dedicated senior talent, but it stays out of reach with an anonymous rented seat. As a result, the access decision is where that difference stops being philosophical. Instead, it becomes something your security review can pass.</p>



<h3 class="wp-block-heading">Dedicated is a security property, not just a staffing one</h3>



<p class="wp-block-paragraph">People usually sell dedicated versus rented as a quality and continuity argument. For example, you get the same senior talent, they learn your codebase, and they stay. That is true, and it matters. Yet it is also a security argument, and that part usually goes unsaid.</p>



<p class="wp-block-paragraph">A named person you can place inside your identity system, scope, review, and offboard on your terms is an accountable person. By contrast, an interchangeable seat is an unaccountable one by construction. When the question is blast radius, accountability is the whole game. The model that keeps the same senior talent embedded in your controls is the same model that lets you answer, cleanly, who touched what.</p>



<p class="wp-block-paragraph">That is the corner the access decision turns on. It is also why the fix for the security blocker is not a longer vendor questionnaire. The fix is a different kind of talent relationship.</p>



<h3 class="wp-block-heading">Where secure nearshore software development actually starts</h3>



<p class="wp-block-paragraph">If your nearshore plan keeps stalling in the same spot, look closely at what is actually stuck. It is rarely a doubt about skill. It is the access grant. And it is the quiet knowledge that a rotating seat leaves you unable to bound or prove that access.</p>



<p class="wp-block-paragraph">That problem has a solution. Instead, add senior talent the way you would add a trusted teammate. First, bring them in through your identity. Then scope them to the work, and put them inside your review gates. Give yourself a footprint you can audit, and a removal of access you can prove. Finally, decide the access question first, and decide it on purpose, so it stops deciding your roadmap by default.</p>



<p class="wp-block-paragraph">At Abstra, that is how dedicated senior talent is meant to work. A named person embedded in your controls, not a seat rented from a pool. If the access decision is the thing holding up your roadmap, start there.</p>



<h3 class="wp-block-heading">FAQ </h3>



<ul class="wp-block-list">
<li><strong>Is the main risk of nearshore development about code quality?</strong>&nbsp;Usually not. You can verify capability up front. The harder risk gets less attention, and it is access. It covers what an outside tech professional can see and reach once you grant the repository, the pipeline, the secrets, and the data. It also covers whether you can bound and prove that access.</li>



<li><strong>How do you limit the blast radius of an outside tech professional&#8217;s access?</strong>&nbsp;Provision them through your own SSO and identity. Scope access to the specific work, so they get only what they need. Route their work through your process and review gates. Keep a footprint you can audit of what they touched. Revoke access cleanly when the engagement changes.</li>



<li><strong>Why is dedicated senior talent more secure than a staffing pool seat?</strong>&nbsp;A named, dedicated tech professional inside your controls is accountable. You can list their access, and you can prove you removed it. An interchangeable pooled seat is harder to attribute, and it may rotate to another client. That is exactly what makes the question who touched this hard to answer.</li>
</ul>
<p>The post <a href="https://abstra.co/blog/secure-nearshore-software-development-access/">Secure Nearshore Software Development Starts on Day One</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>You Acquired the Tech Debt, Too</title>
		<link>https://abstra.co/blog/nearshore-legacy-modernization-services/</link>
		
		<dc:creator><![CDATA[Abstra Team]]></dc:creator>
		<pubDate>Tue, 25 Aug 2026 20:11:33 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482062000</guid>

					<description><![CDATA[<p>Every acquisition brings a second codebase along with the deal, and most integration plans don't account for it. Pulling existing staff off their current work to fix it rarely holds past a few weeks. What works instead is senior tech talent added specifically for the modernization work, embedded for as long as the integration takes rather than a fixed-scope project.</p>
<p>The post <a href="https://abstra.co/blog/nearshore-legacy-modernization-services/">You Acquired the Tech Debt, Too</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Every acquisition brings the acquirer three things: a product, a customer base, and a second tech team&#8217;s codebase. Deal reviews evaluate the first two closely. The third usually gets a line item, not a plan.</p>



<p class="wp-block-paragraph">That gap follows a predictable pattern. The org charts merge quickly, often within weeks. The codebases don&#8217;t. Each system keeps running as-is, because it&#8217;s already serving real customers. Eventually, someone has to decide: migrate it, wrap it, or maintain both indefinitely. That decision competes directly with everything the tech team was already building.</p>



<h3 class="wp-block-heading">Why the Debt Compounds After a Deal</h3>



<p class="wp-block-paragraph">Technical debt is expensive even without an acquisition involved. Developers spend roughly a third of their time on legacy maintenance instead of new work. Most technology leaders say legacy systems slow down their ability to ship anything new. An acquisition doesn&#8217;t create this dynamic: it delivers a second, fully-formed version of it on the deal&#8217;s timeline, instead of letting it build up gradually. As one CTO framework for technical debt puts it: this is why tech teams stay frustrated, and why delivery stays behind. Because it does.</p>



<h3 class="wp-block-heading">Why the Common Fix Doesn&#8217;t Work</h3>



<p class="wp-block-paragraph">The default response is to solve this with existing staff. Pull two or three senior tech professionals off their current work. Form an integration team. Give it a fixed window, usually a quarter. In practice, the window rarely holds. Whatever those professionals were already responsible for doesn&#8217;t pause. As a result, the integration team shrinks as people get pulled back to their original work, and the underlying conflict between the two systems doesn&#8217;t disappear; it just becomes less visible.</p>



<h3 class="wp-block-heading">What Works Instead</h3>



<p class="wp-block-paragraph">The alternative isn&#8217;t a full replatforming project, and it isn&#8217;t simply adding headcount. It&#8217;s adding senior tech talent scoped specifically to the modernization work. These professionals take on the job of learning an unfamiliar codebase, mapping what can be migrated versus what needs to be maintained in place, and executing that plan, without pulling anyone off the work already underway. As a result, the core team stays focused on what the business already expects, while the integration work gets dedicated attention instead of borrowed attention.</p>



<h3 class="wp-block-heading">Where Abstra Fits</h3>



<p class="wp-block-paragraph">This is the specific problem Abstra&#8217;s Dedicated Teams model is built to address. In this model, senior tech talent doesn&#8217;t rotate through as an outsourced project crew; it becomes, in Abstra&#8217;s own description of the model, &#8220;complete software teams that become integral to your organization.&#8221; For a modernization problem, that distinction is the whole point: a team that&#8217;s still accountable in month nine of an integration, not just at kickoff.</p>



<p class="wp-block-paragraph">That kind of sustained, judgment-heavy engagement is Abstra&#8217;s actual track record, not just a claim about the model. Abstra has supported clients&#8217; technical work from their earliest stages onward: cloud-native architecture and platform reliability work carried through multiple growth phases, continuing well past any onboarding period. Abstra&#8217;s own leadership describes the approach as building solutions for problems that don&#8217;t come with a fixed scope: &#8220;From day one, our goal has been to address real-world challenges with flexible, forward-thinking solutions.&#8221; A post-acquisition integration is exactly that kind of problem: it doesn&#8217;t arrive with a fixed scope, and it doesn&#8217;t end on a predictable date.</p>



<p class="wp-block-paragraph">Leadership accountability matters here too. Abstra&#8217;s leadership team has 15+ years bridging U.S. and LATAM markets, which means the person accountable for how the integration is actually going is leadership with direct visibility into the engagement, not a rotating account manager reading status updates back to you.</p>



<p class="wp-block-paragraph">None of that replaces the judgment call every company still has to make for itself: what to migrate, what to wrap, what to leave alone. What it changes is who&#8217;s available to make that call alongside you, for as long as it takes to get right.</p>



<h3 class="wp-block-heading">The Takeaway</h3>



<p class="wp-block-paragraph">If a tech org is still negotiating which of two systems wins at every planning meeting, months after a deal closed, that&#8217;s not a scheduling issue. It&#8217;s a staffing gap with a specific shape. It&#8217;s worth addressing directly, rather than working around it indefinitely.</p>



<h2 class="wp-block-heading">Frequently Asked Questions</h2>



<h3 class="wp-block-heading">What happens to a company&#8217;s tech stack after an acquisition?</h3>



<p class="wp-block-paragraph">It usually keeps running as-is, separate from the buyer&#8217;s own systems, until someone decides to migrate, wrap, or maintain it.</p>



<h3 class="wp-block-heading">Why does technical debt increase after a merger or acquisition?</h3>



<p class="wp-block-paragraph">The deal adds a second, fully-formed codebase on top of whatever debt already existed, all at once instead of gradually.</p>



<h3 class="wp-block-heading">Should internal staff handle post-acquisition system integration?</h3>



<p class="wp-block-paragraph">Rarely works well. Their original responsibilities don&#8217;t pause, so the integration effort loses people back to the core workload within weeks.</p>



<h3 class="wp-block-heading">What are nearshore legacy modernization services?</h3>



<p class="wp-block-paragraph">Technical support from a nearby time zone, typically Latin America, focused on migrating, wrapping, or maintaining outdated systems.</p>



<h3 class="wp-block-heading">How long does post-acquisition system integration usually take?</h3>



<p class="wp-block-paragraph">Longer than the one-quarter window most companies budget for it. There&#8217;s no fixed timeline.</p>



<h3 class="wp-block-heading">How is a dedicated nearshore team different from a project-based vendor?</h3>



<p class="wp-block-paragraph">A dedicated team stays embedded for as long as the work takes. A project-based vendor leaves once a fixed scope is delivered.</p>



<p class="wp-block-paragraph"></p>
<p>The post <a href="https://abstra.co/blog/nearshore-legacy-modernization-services/">You Acquired the Tech Debt, Too</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>The Hire Isn&#8217;t Finished When They Sign</title>
		<link>https://abstra.co/blog/hire-senior-talent-faster/</link>
		
		<dc:creator><![CDATA[Abstra Team]]></dc:creator>
		<pubDate>Wed, 19 Aug 2026 18:03:22 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Business]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482061993</guid>

					<description><![CDATA[<p>Teams that hire senior talent faster often just shorten time-to-offer, not time-to-contribution. Ramp-up still averages three to nine months. Track when a hire starts contributing, not when they sign.</p>
<p>The post <a href="https://abstra.co/blog/hire-senior-talent-faster/">The Hire Isn&#8217;t Finished When They Sign</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">Something quiet gets missed in most hiring conversations. The day an offer letter is signed gets treated like the finish line. It almost never is.</p>



<p class="wp-block-paragraph">Ask a VP of Engineering how they hire senior talent faster than they did last year, and they will usually point to a number: forty-five days, thirty days, maybe less. That number describes how long it took to get a signature. It says very little about how long it took to get help.</p>



<h3 class="wp-block-heading">Two clocks, and only one gets watched</h3>



<p class="wp-block-paragraph">This matters more than it sounds like it should. A scaleup engineering team lives and dies by delivery, not by headcount. So the honest question is not &#8220;how fast can we hire senior talent.&#8221; It is &#8220;how fast does a new hire start moving the roadmap.&#8221; Those are two different clocks, and most companies are only watching one of them.</p>



<p class="wp-block-paragraph">The result is a strange kind of progress that does not feel like progress. Recruiting closes the role. The dashboard turns green. And the team is still short-handed for months, because the clock that matters, time to real contribution, was never being tracked.</p>



<h3 class="wp-block-heading">The months nobody puts on a dashboard</h3>



<p class="wp-block-paragraph">The gap shows up the same way almost every time. A senior role opens. Recruiting moves quickly, interviews happen, an offer goes out, and everyone treats the search as solved. Then the new hire joins, and the team quietly slows down again while that person learns the codebase, the deploy process, and who to ask when something breaks.</p>



<p class="wp-block-paragraph">This is not a personal failing of the hire. It is math. A survey of more than eighty engineers and engineering managers, run by the engineering tooling company Swimm, found it takes three to nine months on average to fully ramp up in a new role, with a meaningful share of companies reporting closer to a year. Signing fast does not change that curve. It just moves the slow part later, where it is easier to miss.</p>



<p class="wp-block-paragraph">For a VP of Engineering staring at a roadmap that is already behind, that hidden lag is the real cost. Not the salary. Not the recruiting fee. The quarter that slips while a &#8220;fast hire&#8221; is still learning where things live.</p>



<h3 class="wp-block-heading">What scaling looks like to the people who lived it</h3>



<p class="wp-block-paragraph">One partner we have worked with put this plainly. &#8220;Highly recommend Abstra for their ability to help us scale with strong technical resources across many of our scrum teams,&#8221; said their Director of Engineering. The word doing the work in that sentence is &#8220;scale,&#8221; not &#8220;staff.&#8221; Scaling a team means the new person is contributing inside the team&#8217;s actual rhythm, not sitting beside it while everyone waits.</p>



<h3 class="wp-block-heading">Closing the gap, not just filling the seat</h3>



<p class="wp-block-paragraph">The anti-pain here is not slow hiring. It is hiring that looks fast on paper and slow in practice. A process built to optimize time-to-offer will keep producing that gap, no matter how quickly recruiting moves.</p>



<p class="wp-block-paragraph">This is the part of the work we care about at Abstra. Our&nbsp;<a href="https://abstra.co/services-staff-augmentation/">staff augmentation</a>&nbsp;model exists because senior professionals who have already worked inside real product teams, many of them based in Latin America, do not need three to nine months to get moving. Our&nbsp;<a href="https://abstra.co/services-dedicated-teams/">dedicated teams</a>&nbsp;model goes further: it is not a supplement bolted onto a struggling roadmap. It is built to join the team&#8217;s existing workflow from week one, instead of adjusting to it later. In practice, that is the difference between filling a seat and closing a gap. Our&nbsp;<a href="https://abstra.co/our-clients/">clients</a>&nbsp;tend to describe it the same way that partner did: not faster hiring, but faster scaling.</p>



<h3 class="wp-block-heading">The question worth sitting with</h3>



<p class="wp-block-paragraph">None of this means time-to-offer is meaningless. It just is not the number that tells you whether delivery is back on track. The next time a senior role opens, and the pressure is on to hire senior talent faster, it is worth asking a second question alongside the first: when this person joins, how long until they are part of the work, not just on the team?</p>



<h3 class="wp-block-heading">Questions people ask about hiring senior talent faster</h3>



<ul class="wp-block-list">
<li><strong>What does it mean to hire senior talent faster without sacrificing quality?</strong>&nbsp;It means measuring speed by how quickly a new hire becomes a contributing part of the team, not only by how quickly an offer is signed. A fast signature paired with a slow ramp-up is not fast hiring.</li>



<li><strong>How long does it typically take a new senior hire to ramp up?</strong>&nbsp;Industry data points to three to nine months on average, based on a survey of over eighty engineers and engineering managers conducted by Swimm, with some companies reporting closer to a year.</li>



<li><strong>What is the difference between staff augmentation and dedicated teams?</strong>&nbsp;Staff augmentation adds vetted individual professionals into an existing team quickly. Dedicated teams build a full team structured around a specific product area, integrated into the client&#8217;s workflow from the start.</li>



<li><strong>Does nearshore hiring reduce ramp-up time?</strong>&nbsp;It can, when the professionals joining have already worked inside comparable product teams and processes. The advantage is not speed to signature. It is fewer surprises once the person starts.</li>
</ul>
<p>The post <a href="https://abstra.co/blog/hire-senior-talent-faster/">The Hire Isn&#8217;t Finished When They Sign</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>The Same Care You Give a Tool, Give the Team</title>
		<link>https://abstra.co/blog/secure-nearshore-development-partner/</link>
		
		<dc:creator><![CDATA[Abstra Team]]></dc:creator>
		<pubDate>Tue, 04 Aug 2026 21:06:05 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Partnership]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482061986</guid>

					<description><![CDATA[<p>You review every SaaS tool, then give an outside team the same access with less thought. A secure nearshore development partner makes access, IP, and accountability part of the work, backed by a short line to US-bred leadership.</p>
<p>The post <a href="https://abstra.co/blog/secure-nearshore-development-partner/">The Same Care You Give a Tool, Give the Team</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<p class="wp-block-paragraph">There is a certain care most teams bring to the tools they choose. The new software gets a few honest questions. Where does the data live. Who can reach it. What happens if it breaks. We give a tool that attention because it will sit close to our work. The people we invite into the same work deserve at least as much thought. That is the quiet idea behind choosing a&nbsp;<strong>secure nearshore development partner</strong>.</p>



<p class="wp-block-paragraph">The people who join a team shape it more than any tool does. They see the codebase, the trade-offs, the way the work really happens. So the care we put into choosing them matters. And it is worth putting into a partner too, not only into software.</p>



<h3 class="wp-block-heading">The attention a tool earns</h3>



<p class="wp-block-paragraph">Think about the care you bring to a SaaS vendor. You ask where the data lives. You look at access controls. You read the incident history. You would not put an unknown tool into production on the strength of a demo, because it earns a closer look first.</p>



<p class="wp-block-paragraph">An outside team deserves that same closer look, and often it gets less. The choice comes down to a rate, a few resumes, and a good conversation. Yet those people become a third party with real access. And third parties are where risk is quietly moving. Third-party involvement in breaches doubled year over year, to 30% of all incidents. Those same breaches cost about $4.91M on average. And they take roughly 26 days longer to detect (<a href="https://www.ibm.com/reports/data-breach">IBM Cost of a Data Breach 2025</a>). Those are facts. So here is the judgment. The outside team writing your code is a third party too. And it is worth the same care.</p>



<h3 class="wp-block-heading">Why the quiet risk stays quiet</h3>



<p class="wp-block-paragraph">If you work in a regulated industry, an auditor forces the question for you. HIPAA, PCI, or SOC 2 makes you write down who touches what. But most companies are not there yet. So nothing forces the conversation, and it becomes a feeling instead of a checklist. A founder wants to hire nearshore, then stalls, and cannot quite say why.</p>



<p class="wp-block-paragraph">The why is usually trust, not price or time zone. Inviting people you have not met into your codebase feels like a security decision you have no framework for. So it keeps waiting. A&nbsp;<strong>secure nearshore development partner</strong>&nbsp;should offer you that framework, and then meet it comfortably.</p>



<h3 class="wp-block-heading">The safeguard is a person you can reach</h3>



<p class="wp-block-paragraph">A checklist helps. But what protects you is accountability. And accountability has a name and a phone number. One of our clients, Tyler Meadlin, CTO of Certiverse,&nbsp;<a href="https://abstra.co/our-clients/">put it this way</a>: &#8220;An invaluable partner for our startup. Abstra&#8217;s expertise and technical exposure have been instrumental in helping us navigate challenges and develop scalable solutions.&#8221;</p>



<p class="wp-block-paragraph">Notice the word partner. Not vendor, not bench. When something gets hard, you want a direct line to people who own the outcome with you, not a ticket queue. That is the difference between a firm you buy from and a partner who answers.</p>



<h3 class="wp-block-heading">What that care looks like in practice</h3>



<p class="wp-block-paragraph">So this is where Abstra takes a position. Trust is not a slide. It is a structure. US-bred founders built Abstra. So your escalation path stays short, and the people who set the standard stay within reach, not an ocean and three layers away. See how the team&nbsp;<a href="https://abstra.co/blog/abstra-cybersecurity-approach/">treats security as judgment, not just tooling</a>. Our cybersecurity lead describes the real safeguard as a person who brings &#8220;rigor, anticipation, and decision-making under pressure.&#8221;</p>



<p class="wp-block-paragraph">That structure runs quietly through the work. Least-privilege access, so people hold only what the task needs. Clear IP assignment, so what your team builds stays yours. Careful offboarding, so access ends when a role ends. And it covers senior talent across every role a product needs. The engineers, yes. And the QA, DevOps, data, and product people too. We choose them all for judgment first. You can meet the model behind it in our&nbsp;<a href="https://abstra.co/services-dedicated-teams/">dedicated Latin America teams</a>, built to stay and to stand behind their work.</p>



<h3 class="wp-block-heading">Why the culture matters more than the clause</h3>



<p class="wp-block-paragraph">None of this holds because a contract says so. It holds because of how the team thinks. Our&nbsp;<a href="https://abstra.co/abstra-manifesto/">manifesto</a>&nbsp;puts it plainly: &#8220;Security is part of the job. Do not be the reason something leaks.&#8221; Beside it sits another line the whole company lives by: &#8220;We are not vendors, we are partners. Act like it is your company.&#8221; That culture grew from the start. Two friends from Paraguay, trained in the United States, built Abstra in 2008. They wanted to connect careful people across borders, and keep the work human. So when the people in your codebase treat your data as their own, a security review stops being a gate you enforce. It becomes an instinct you share.</p>



<h3 class="wp-block-heading">A gentle question, worth asking first</h3>



<p class="wp-block-paragraph">Most security trouble now arrives quietly. It comes through a third party with access, not through a dramatic break-in. The outside team in your work is that third party. So the question worth sitting with is a kind one. Would you give these people the same careful look you give every tool? Would they welcome it? And who picks up the phone if something ever goes wrong? Choose a&nbsp;<strong>secure nearshore development partner</strong>&nbsp;you can hold close, and hold accountable. Then the trust stops being a weight you carry alone. See how a&nbsp;<a href="https://abstra.co/services-dedicated-teams/">dedicated Latin America team</a>&nbsp;earns it.</p>



<h3 class="wp-block-heading">FAQ</h3>



<ul class="wp-block-list">
<li><strong>What makes a secure nearshore development partner different from low-cost staff augmentation?</strong>&nbsp;A secure nearshore development partner treats access, IP, and accountability as part of the work, not an afterthought. You get senior talent chosen for judgment, least-privilege access, clear IP assignment, and a short line to people who own the outcome with you.</li>



<li><strong>We are not in a regulated industry. Do we still need a security review of an outside team?</strong>&nbsp;Yes. Regulation forces the question, but the risk exists either way. Third-party involvement in breaches reached 30% of incidents in 2025, so the outside team with access to your systems is worth the same scrutiny you give a software vendor.</li>



<li><strong>Does a nearshore partner only provide engineers?</strong>&nbsp;No. Abstra places senior talent across the roles a product needs, including QA, DevOps, data, product, and design, not only engineers. Each is vetted for judgment and works in your time zone.</li>



<li><strong>How does US-bred leadership change the security picture?</strong>&nbsp;It shortens the distance between you and the people who set the standard. When a hard call comes up, you reach a decision-maker who owns it with you, instead of a queue.</li>
</ul>
<p>The post <a href="https://abstra.co/blog/secure-nearshore-development-partner/">The Same Care You Give a Tool, Give the Team</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
		<item>
		<title>What a Real AI Reskilling Strategy Gets Right</title>
		<link>https://abstra.co/blog/ai-reskilling-strategy/</link>
		
		<dc:creator><![CDATA[Abstra Team]]></dc:creator>
		<pubDate>Fri, 31 Jul 2026 17:55:24 +0000</pubDate>
				<category><![CDATA[Blog]]></category>
		<category><![CDATA[Business]]></category>
		<category><![CDATA[Technology]]></category>
		<guid isPermaLink="false">https://abstra.co/?p=9482061981</guid>

					<description><![CDATA[<p>One global retailer built an AI reskilling strategy around its customer service AI assistant. It retrained about 8,500 call center workers into design consultants instead of cutting the team. That move grew into a channel worth over a billion euros a year. The lesson is not that AI avoids all layoffs, but that a deliberate reskilling strategy can turn automation into new revenue instead of just lower cost.</p>
<p>The post <a href="https://abstra.co/blog/ai-reskilling-strategy/">What a Real AI Reskilling Strategy Gets Right</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></description>
										<content:encoded><![CDATA[
<div class="et_pb_section et_pb_section_0 et_section_regular" >
				
				
				
				
				
				
				<div class="et_pb_row et_pb_row_0">
				<div class="et_pb_column et_pb_column_4_4 et_pb_column_0  et_pb_css_mix_blend_mode_passthrough et-last-child">
				
				
				
				
				<div class="et_pb_module et_pb_text et_pb_text_0  et_pb_text_align_left et_pb_bg_layout_light">
				
				
				
				
				<div class="et_pb_text_inner"><!-- divi:paragraph -->
<p>Most conversations about AI and jobs start from the same fear. Automation is a trade: machines for people, and someone always loses. One well-documented case from a global home furnishings retailer complicates that story.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>A few years ago, the company introduced an AI assistant to handle customer service. In its first two years, the assistant could resolve 47 percent of customer questions on its own. Today that figure is 74 percent, according to public reporting on the program. A support function got faster and cheaper almost overnight. That part of the story is familiar. What the company did next is the part worth building an AI reskilling strategy around, not just admiring.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading">The choice most companies skip</h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>When a chatbot starts absorbing routine questions, the obvious move is to shrink the team that used to answer them. This company took a different path. It looked at the roughly 8,500 call center employees whose workload had shrunk. Then it asked a harder question: what were customers still calling about, and could a person do that better than a bot?</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>The answer was design. Customers who got past the basic questions were usually trying to plan a kitchen, a closet, or a whole room. Those problems need judgment, taste, and a back-and-forth a bot cannot yet offer. So the company retrained those 8,500 workers as remote design consultants instead of letting them go. The retraining took about two years for the original group. It now takes five to six weeks for new hires.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>That is a real cost. Training at that scale is not free, and it does not pay off in a single quarter. Leaders under pressure to show fast AI returns rarely choose it. This company did. That choice was a judgment about the business, not a guaranteed formula for every company.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading">What the numbers earned</h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>The results are specific enough to check. Reskilled workers now staff a remote-sales channel. That channel has been the company&#8217;s fastest-growing sales line for three straight years. It grew 15 to 20 percent annually and brought in the equivalent of over a billion euros last fiscal year, up from the year before. The company also reports its customer happiness score has climbed to 89 percent, up from 60 percent before the AI assistant existed.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>None of that erases the layoffs elsewhere in the broader business. Public reporting notes the company&#8217;s corporate structure cut hundreds of office-based roles the same year, unrelated to this program. The company did not attribute those cuts to AI. They also left the remote-sales team untouched. The honest version of this story is not &#8220;this company never cuts jobs.&#8221; It is narrower and more useful. The specific team affected by this AI rollout kept their jobs. They moved into higher-value work and helped build a channel worth over a billion euros a year.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading">Why this matters beyond one company</h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>We see a version of the same fear in almost every conversation about adding AI to an engineering or support function. Leadership wants the efficiency, but the team hears &#8220;fewer of us.&#8221; That fear is not irrational. It is also not the only outcome available.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>This is our position, not just an observation about one retailer. When a tool makes a piece of work faster or cheaper, demand for that work has historically grown rather than shrunk. Economists have called this the Jevons paradox since the 1800s. We think the pattern holds for engineering now. Narrow tasks disappear. Judgment work tends to grow instead: deciding what an AI system should own, reviewing what it produces, staying accountable for what ships. We laid out that argument in more depth in&nbsp;<a href="https://abstra.co/blog/ai-and-jobs-paradox/">AI and jobs: what the Jevons paradox reveals about the future of work</a>.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading">What the company got right</h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>This case works because someone did the harder analysis before making the harder promise. They looked at what the AI could not do well. They found the higher-value work hiding behind that gap, and built a training path into it. That is a design decision, not a lucky break. It is the kind of decision that benefits from technical partners who have built AI-augmented systems before, not just deployed a chatbot and hoped.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>Good AI implementation and good nearshore engineering practice overlap more than people expect. Both depend on scoping the work honestly. What should the machine own? What should stay human? What does the team need to learn to do the second part well? We wrote about that overlap in&nbsp;<a href="https://abstra.co/blog/nearshore-software-development-ai-era/">nearshore in the AI era: the part that stays human</a>. We covered the judgment AI still cannot replace in&nbsp;<a href="https://abstra.co/blog/human-on-the-loop-ai/">the rise of agentic AI and why human-on-the-loop is the new standard</a>.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading">A question worth sitting with before the next AI rollout</h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>This story is not a template anyone can copy exactly. Few companies have the balance sheet or the adjacent revenue opportunity that made a two-year retraining program worth the wait. But the underlying question travels well. When a piece of work gets automated, does anyone in the room ask what the freed-up people could do instead? Or does the conversation stop at headcount?</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>So maybe the useful question is not whether AI will change a given role. It probably will. It is whether anyone has looked hard enough at what&#8217;s left over to find where the people go next.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>Most conversations about AI and jobs start from the same fear: that automation is a trade, machines for people, and someone always loses. One well-documented case from a global home furnishings retailer complicates that story.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>A few years ago, the company introduced an AI assistant to handle customer service. In its first two years, the assistant could resolve 47 percent of customer questions on its own. Today that figure is 74 percent, according to public reporting on the program. A support function got faster and cheaper almost overnight. That part of the story is familiar. What the company did next is the part worth building an AI reskilling strategy around, not just admiring.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading"><strong>The choice most companies skip</strong></h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>When a chatbot starts absorbing routine questions, the obvious move is to shrink the team that used to answer them. This company looked at the roughly 8,500 call center employees whose workload had shrunk and asked a different question: what were customers still calling about, and could a person do that better than a bot?</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>The answer was design. Customers who got past the basic questions were usually trying to plan a kitchen, a closet, a whole room, problems that need judgment, taste, and a back-and-forth a bot cannot yet offer. So the company retrained those 8,500 workers as remote design consultants instead of letting them go. The retraining took about two years for the original group and now takes five to six weeks for new hires.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>That is a real cost. Training at that scale is not free, and it does not pay off in a single quarter. Leaders under pressure to show fast AI returns rarely choose it. This company did, and the fact that it did is a judgment about the business, not a guaranteed formula for every company.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading"><strong>What the numbers earned</strong></h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>The results are specific enough to check. The remote-sales channel built around those reskilled workers has been the company&#8217;s fastest-growing sales channel for three straight years, growing 15 to 20 percent annually, and brought in the equivalent of over a billion euros in the last fiscal year, up from the year before. The company also reports its customer happiness score has climbed to 89 percent, up from 60 percent before the AI assistant existed.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>None of that erases the fact that the broader business has had layoffs elsewhere. Public reporting notes the company&#8217;s corporate structure cut hundreds of office-based roles the same year, unrelated to this program. Those cuts were not tied to AI, and they did not touch the remote-sales team. The honest version of this story is not &#8220;this company never cuts jobs.&#8221; It is narrower and more useful: the specific team affected by this specific AI rollout kept their jobs, moved into higher-value work, and helped build a channel worth over a billion euros a year.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading"><strong>Why this matters beyond one company</strong></h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>At Abstra, we see a version of the same fear in almost every conversation about adding AI to an engineering or support function: leadership wants the efficiency, but the team hears &#8220;fewer of us.&#8221; That fear is not irrational. It is also not the only outcome available.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>This is our position, not just an observation about one retailer. When a tool makes a piece of work faster or cheaper, demand for that work has historically grown rather than shrunk. Economists have called this the Jevons paradox since the 1800s, and we think the pattern holds for engineering now. The narrow tasks disappear. The roles built on judgment, deciding what an AI system should own, reviewing what it produces, staying accountable for what ships, tend to grow instead. We laid out that argument in more depth in&nbsp;<a href="https://abstra.co/blog/ai-and-jobs-paradox/">AI and jobs: what the Jevons paradox reveals about the future of work</a>.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>This case works because someone did the harder analysis before making the harder promise. They looked at what the AI could not do well, found the adjacent, higher-value work hiding behind that gap, and built a training path into it. That is a design decision, not a lucky break, and it is the kind of decision that benefits from technical partners who have built AI-augmented systems before, not just deployed a chatbot and hoped.</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>This is where good AI implementation and good nearshore engineering practice overlap more than people expect. Both depend on scoping the work honestly: what should the machine own, what should stay human, and what does the team need to learn to do the second part well. We wrote about that overlap directly in&nbsp;<a href="https://abstra.co/blog/nearshore-software-development-ai-era/">nearshore in the AI era: the part that stays human</a>, and about the judgment AI still cannot replace in&nbsp;<a href="https://abstra.co/blog/human-on-the-loop-ai/">the rise of agentic AI and why human-on-the-loop is the new standard</a>.</p>
<!-- /divi:paragraph -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading"><strong>A question worth sitting with before the next AI rollout</strong></h3>
<!-- /divi:heading -->

<!-- divi:paragraph -->
<p>This story is not a template anyone can copy exactly. Few companies have the balance sheet, the timeline, or the adjacent revenue opportunity that made a two-year retraining program worth the wait. But the underlying question travels well: when a piece of work gets automated, does anyone in the room ask what the freed-up people could do instead, or does the conversation stop at headcount?</p>
<!-- /divi:paragraph -->

<!-- divi:paragraph -->
<p>So maybe the useful question is not whether AI will change a given role. It probably will. It is whether anyone has looked hard enough at what&#8217;s left over to find where the people go next.</p>
<!-- /divi:paragraph -->

<!-- divi:separator -->
<hr class="wp-block-separator has-alpha-channel-opacity"/>
<!-- /divi:separator -->

<!-- divi:heading {"level":3} -->
<h3 class="wp-block-heading">FAQ</h3>
<!-- /divi:heading -->

<!-- divi:list -->
<ul class="wp-block-list"><!-- divi:list-item -->
<li><strong>What is an AI reskilling strategy?</strong>&nbsp;An AI reskilling strategy is a deliberate plan to retrain employees whose work is partly automated into new roles that use the capacity AI freed up, rather than reducing headcount. One well-documented retailer used this approach when its AI assistant began handling routine customer service questions.</li>
<!-- /divi:list-item -->

<!-- divi:list-item -->
<li><strong>Did that company really avoid layoffs when it introduced AI?</strong>&nbsp;For the specific team affected, yes. The company retrained around 8,500 call center employees into remote design consultant roles instead of cutting those positions, according to public reporting. The broader business has had separate corporate layoffs since, but those were not attributed to AI and did not affect the remote-sales team.</li>
<!-- /divi:list-item -->

<!-- divi:list-item -->
<li><strong>How much revenue did this AI reskilling strategy generate?</strong>&nbsp;Public reporting puts the remote-sales channel, staffed largely by the reskilled customer service team, at the equivalent of over a billion euros in the most recent fiscal year, making it the company&#8217;s fastest-growing sales channel for three consecutive years.</li>
<!-- /divi:list-item -->

<!-- divi:list-item -->
<li><strong>Can smaller companies apply the same AI reskilling strategy?</strong>&nbsp;The core idea scales down even if the exact program does not. Any company automating part of a role can ask what work is left over that needs human judgment, and whether existing employees could be trained into it, rather than defaulting to reducing the team.</li>
<!-- /divi:list-item -->

<!-- divi:list-item -->
<li><strong>What should a company look for in an AI implementation partner?</strong>&nbsp;A partner who can help identify what should stay human before deciding what to automate, not one who treats every AI rollout as a cost-cutting exercise by default.</li>
<!-- /divi:list-item --></ul>
<!-- /divi:list --></div>
			</div>
			</div>
				
				
				
				
			</div>
				
				
			</div>
<p>The post <a href="https://abstra.co/blog/ai-reskilling-strategy/">What a Real AI Reskilling Strategy Gets Right</a> appeared first on <a href="https://abstra.co">Abstra</a>.</p>
]]></content:encoded>
					
		
		
			</item>
	</channel>
</rss>
