Hey, I'm Adi

I'm a data scientist turned AI engineer who builds performant AI agents and ships them with data science driven evaluations. I've spent 6 years working on LLMs, mostly deploying them in US healthcare.

This is my digital garden where I nerd out about AI agent evaluation, memory systems, retrieval architectures, and the intersection of AI and healthcare. Expect technical deep dives, half-baked ideas, and lots of learning in public.

Featured

• Evals vs Benchmarks vs Verifiers

  13 Jul, 2026

  A gentle intro to how these are similar but different.

• We Need Owners

  6 Jul, 2026

  AI has a deep trust problem because it cannot be held responsible for its own output like humans can be.

• Big-O of AI Productivity

  23 Apr, 2026

  The nuanced lessons I learnt in my journey to "10X" myself with AI