garva manai.
HomeEssaysProduction AISAAL FrameworkNewsletterConnect
// est. 2021 · writing from bengaluru & sf

Field notes on shipping
actually useful AI.

I help engineering teams move language models out of demos and into revenue. These are the essays, frameworks and post-mortems that came out of that work.

Read latest essaySubscribe to the newsletter
Recent essays
All 24 essays
No. 023

Why your eval set is lying to you.

On the slow decay of golden datasets, and what to do when prod drifts from your bench.

Mar 18, 2026 · 14 min
No. 022

Retrieval is a product surface.

Treating your RAG layer as a UI decision, not a backend tuning problem.

Mar 02, 2026 · 11 min
No. 021

The SAAL Framework, v0.4.

Scope · Anchor · Audit · Loop — an updated diagram with notes on where teams stall.

Feb 14, 2026 · 18 min
No. 020

A handoff document I wish I'd had.

Field-tested template for transferring an LLM system from prototype to platform team.

Jan 29, 2026 · 9 min
No. 019

Three classes of LLM bug.

Brittle prompt, leaky context, drifting world — and the eval each one demands.

Jan 09, 2026 · 13 min
portrait · garva manai
About

I write what I learn in production.

Eight years building ML systems — three of them spent prying language models into legacy enterprise stacks. I've shipped customer-facing LLM products at two Series-B startups and one very large bank.

This site is the public-facing notebook. Essays are long arguments, field notes are short ones, and the SAAL Framework is the diagram I draw at every kickoff. Everything here is freely re-usable — attribution appreciated, citation optional.

The newsletter

One careful essay, every other Sunday.

No threads, no take-laundering. 4,800 engineers and PMs read it to think more clearly about LLMs in production.

// archive at /newsletter · unsubscribe in one click