QA Architect & Lead SDET with 10+ years engineering AI-powered testing systems at scale. Currently driving quality at Mendix — where releases ship without fear.
Python · Playwright · AI Automation · CI/CD · Cloud-Native · Microservices
I started where most engineers start — writing code, shipping features, wondering why things kept breaking in ways nobody anticipated. That question became an obsession.
Over 10+ years across SaaS platforms, telecom infrastructure, and cloud-native microservices, I've moved from writing tests to designing the systems that make entire teams confident in every release.
Today I work as a freelance QA Architect and Lead SDET — embedding quality into products from day one, building automation that evolves with the application, and introducing AI where it genuinely reduces noise rather than adding hype.
I don't take on projects where I'll be writing tickets. I take on projects where I can change how quality works.
// how I engage
Design scalable UI, API, and E2E frameworks from scratch — built to survive real teams and real velocity.
Self-healing selectors, AI-assisted test generation, and intelligent failure triage — less maintenance, more signal.
Testing embedded into every pipeline stage — not as a gate, but as a continuous confidence signal.
Cloud-native microservices, telecom infrastructure, device firmware — I've tested systems where failure isn't an option.
Quality KPIs, coverage dashboards, and real-time failure intelligence — decisions driven by data, not gut feel.
Define testing standards, mentor engineers, and embed a quality culture across distributed teams.
Every metric here reflects real engineering decisions with real business consequences — not vanity numbers.
CI/CD-integrated testing cut deployment time across Mendix cloud-native pipelines.
Built end-to-end, API, and UI test suites covering critical paths across microservices.
Self-healing mechanisms and observability-driven analysis cut test instability.
Continuous quality engineering across telecom, SaaS, and cloud-native platforms.
Defined and drove QA standards, mentoring engineers across distributed Mendix teams.
DOCSIS, TCP/IP, firmware validation on millions of broadband devices at Liberty Global.
"Most teams don't have a testing problem. They have a confidence problem in their releases."
This is my mental model. Every system I design moves quality from reactive to predictive — from manual to self-governing.
AI risk analysis predicts which tests to run, reducing execution time by targeting high-impact areas.
Self-healing selectors and adaptive locators reduce maintenance to near zero — tests evolve with the app.
Observability integration means failures surface with context, not just stack traces — faster root cause.
Not just tools — systems thinking across automation, AI, infrastructure, and observability.
Each project was born from a real problem. Each solution is still running in production.
Teams wasted days writing boilerplate test cases for every new feature — slow, repetitive, and error-prone.
Built an internal AI agent that analyzes API specs, user flows, and code diffs to auto-generate meaningful test scenarios — reducing test authoring time by 70%.
70% faster test authoring · Coverage up 40% · Zero test debt accumulation
UI tests were breaking every sprint due to minor frontend changes — team spent more time fixing tests than writing new ones.
Implemented an adaptive locator system that detects broken selectors and automatically heals them using AI-powered element matching and DOM analysis.
85% reduction in test maintenance · Zero flaky blocks in CI · Engineers focus on value
Microservices were breaking each other silently — API contract violations only discovered in production.
Designed a contract-testing layer with automated spec validation, schema drift detection, and inter-service regression testing integrated into every PR pipeline.
100% API contract coverage · Caught 3 critical production regressions pre-release
Testing firmware across 10+ broadband device models (DOCSIS, Wi-Fi, TCP/IP) was manual, inconsistent, and blocked release cycles.
Extended Boardfarm Python framework to support automated firmware validation, protocol-level testing, and parallel device execution across Arris, Compal, Hitron, Ubee, Samsung hardware.
10x test throughput · Uniform validation across 10+ device families · 4.5 years of zero critical escapes
QA metrics lived in spreadsheets. Decisions were gut-based. Nobody knew real release confidence until it was too late.
Built a real-time quality dashboard integrating test results, CI/CD pipeline health, failure trends, and coverage metrics from Datadog and Grafana — giving the entire engineering org a single source of truth.
Quality KPIs visible org-wide · Reduced time-to-insight from days to minutes · Data-driven release decisions
Whether you're a QA engineer leveling up, a team that needs a framework audit, or a leader who wants a clear quality strategy — here's how we can work together beyond a full-time hire.
For QA engineers ready to move beyond writing tests. We work on architecture decisions, career strategy, code reviews, and the mindset shift that separates testers from architects.
I audit your existing test suite — architecture, flakiness patterns, CI integration, coverage gaps. You get a written report with a clear remediation roadmap, prioritised by business impact.
A hands-on half-day or full-day session for engineering teams. Covers AI-driven test generation, self-healing automation, CI/CD quality integration, and observability-driven QA. Remote or on-site.
A focused 90-minute session where I dig into your product, your release process, and your quality pain points — and hand you a concrete strategy: what to build, what to fix, what to stop doing.
Topics I cover
One post a week. Real engineering decisions, specific tools, and the hard lessons from building AI-powered testing systems at scale.
No spam. Unsubscribe any time.
These aren't responsibilities lifted from a job description. They're outcomes from a decade of engineering decisions.
Most engineers start writing tests on day one. I start by understanding the release pipeline, failure patterns, and what confidence actually means for your product.
Self-healing locators, adaptive test structures, and AI-assisted maintenance mean your test suite evolves with your product — not against it.
Testing as a gate is too late. I design quality into every stage of the pipeline so problems surface in seconds, not sprint reviews.
Broadband firmware at Liberty Global. Cloud-native SaaS at Mendix. When stakes are high and failures are public, I've been the engineer who made sure they didn't.
Real AI workflows: test generation from specs, failure analysis from logs, coverage mapping from usage data. Not demos — shipped systems.
Mentoring, standards, frameworks, code reviews — I raise the automation ceiling for every team I join. The knowledge doesn't leave when I do.
I can explain a flaky test to an architect and explain business risk to a CPO. I bridge the gap between technical quality and strategic confidence.
Infosys → Liberty Global → Mendix. Each role was a deliberate step forward in complexity, scope, and impact. I don't plateau.
If you're building something that has to work — let's talk.
From engineers to executives — the pattern is consistent.
"Raju didn't just build automation — he redesigned how we think about quality. Our CI pipelines went from blocking to accelerating. He's the rare QA engineer who operates at architect level from day one."
"Within the first quarter, Raju had mapped our entire test debt and proposed a remediation architecture that reduced flakiness by 80%. He's not a tester — he's a quality engineer who happens to write very good tests."
"What sets Raju apart is his systems thinking. He doesn't patch — he designs. When he introduced AI-assisted test generation, our coverage doubled without doubling the team."
"I've worked with many SDETs. Raju is one of the few who can speak in a technical review, a product meeting, and a board update — and make all three groups act. That's rare."
"He took our test suite from an afterthought to a competitive advantage. Release confidence is now something we actually measure — and we have Raju to thank for that."
"Raju mentored two junior engineers on our team into productive automation contributors in less than 3 months. He doesn't just build systems — he builds teams."
Whether you're hiring, booking a session, or just want to talk quality strategy — I'm easy to reach.