About
From testing software to assuring AI.
AI Assurance Architect and Product Engineering Lead in Abu Dhabi, with 13+ years proving that software, data platforms and now AI systems do what they claim.
The story
I started my career in 2012 testing high-traffic services at Naukri.com. The lesson from those years has stayed with me: quality is a habit, and trust has to be earned with evidence.
Over the next decade I moved from API automation to test frameworks, performance and big data quality, first at Fliplearn and then at Saal.ai in Abu Dhabi. Each step was the same job at a larger scale: prove that a system does what it claims, before customers have to find out that it does not.
When AI moved into production, that question became harder and more important. Today I architect how agentic and LLM systems are evaluated, attacked, guarded, observed and governed. I also build agentic systems myself, which keeps the assurance work practical.
Alongside product work, I help shape solutions for enterprise and government AI assurance programmes across the UAE and GCC, and I mentor a team of engineers growing into this field.
What sets me apart
Three perspectives, one practice.
- A tester’s instincts
- Thirteen years of proving software works, from APIs and mobile apps to data platforms. I start from doubt and finish with evidence.
- A builder’s hands
- I architect and build the systems I assure: agentic AI with LangGraph, CrewAI and MCP, evaluation pipelines and AI platforms.
- A delivery lens
- Roadmaps for executive leadership, solution design for enterprise and government bids, and a team of six engineers I mentor.
Career
Thirteen years, one thread.
AI Assurance Architect & Product Engineering Lead
Saal.ai, Abu Dhabi
- Owns AI assurance and governance across the product portfolio: evaluation criteria, golden datasets and release gates for LLM and agentic products.
- Architects an AI agent reliability and observability platform, and contributes architecture to an agentic AI control plane.
- Builds agentic QA systems with CrewAI, LangGraph, Agno and MCP; leads AI red-teaming and an application security programme.
- Leads technical bids for enterprise and government AI assurance programmes; mentors a team of six engineers.
Senior QA Engineer
Saal.ai, Abu Dhabi
- Scaled automation with Playwright, Selenium, Cypress and Appium in CI/CD, cutting regression from 3 days to under 4 hours.
- Performance work with JMeter and K6 that improved throughput by 35%; over 95% API regression coverage.
Automation QA Engineer
Saal.ai, Abu Dhabi
- Built the first automation framework from scratch and the QA processes that later roles scaled.
Senior QA Engineer
Fliplearn Education, Gurugram
- Introduced shift-left testing that reduced post-release defects by 30%, with under 2% escape rate across 200+ scenarios.
QA Engineer
Info Edge (Naukri.com), Noida
- API automation with Rest Assured and Serenity BDD; nightly Jenkins regression that turned days into hours. Twice named Best QA Engineer.
Recognition
Recognised for the work.
- Best Employee of the YearSaal.ai, 2024 and 2025
- Innovation Hackathon WinnerSaal.ai Innovation Hackathon
- Best QA Engineer, twiceInfo Edge (Naukri.com)
Education & credentials
Foundations.
- B.Tech, Computer Science & EngineeringIILM Academy of Higher Learning, Uttar Pradesh
- UAE Golden Visa holderBased in Abu Dhabi
- Technical writingArticles on big data testing, data quality and TPC-H benchmarking on Medium and Towards Dev, and on this site
Beyond work
What keeps me curious.
Outside work, I make time for reading and learning, for family and community, and for travel and the cultures that come with it. Much of what I learn ends up in my writing and in how I mentor.
What I’m focused on nowWork with me
Want to work together, or compare notes on AI assurance?