AI Test Engineer · Mercedes-Benz R&D India

AI systems need to be tested.
Very few people know how.

I build evaluation and testing infrastructure for LLM systems at Mercedes-Benz R&D. This is where I write about what I learn — for engineers who want to do this properly.

WRITING

21 May 2026 Build RAGAS vs TruLens vs DeepEval — which eval framework should you actually use? Three frameworks, three philosophies. Here's when to reach for each one. →

18 May 2026 Think AI just solved a math problem that stumped humanity for 80 years What the Erdős unit distance breakthrough actually means for AI reasoning. →

14 May 2026 Think Bengaluru hosts 50% of India's AI talent — here's what that actually means The numbers are real but the narrative is incomplete. →

All writing →

Amit Kumar

AI Test Engineer · MBRDI · Bangalore

I came from QA and automation testing. I watched AI features ship without any real evaluation strategy. So I built one — and kept building.

Now I do this full time at Mercedes-Benz R&D India. This site is where I think out loud.

SDET → AI The transition

MBRDI Current role

LLM Eval Core focus

Bengaluru Based in

AI systems need to be tested. Very few people know how.

Amit Kumar

If you're building or testing LLM systems.

AI systems need to be tested.
Very few people know how.