Blog

Notes from the lab.

Research write-ups, build notes and practical guides on agents, evaluation, LLMOps and making AI pay for itself.

Research · 10 min read

Beyond the demo: what it takes to ship AI that lasts

Evaluation, guardrails, cost control and ownership — the four things that separate a prototype from a production system, and how to build each one in from day one.

Subhash Nunna · Oct 2026

Read the post →

Why most AI agents fail in production — and the guardrails that fix them

6 min read

Evaluation

A practical eval stack for LLM apps you can set up in a day

8 min read

AI FinOps

Cutting LLM costs without cutting quality

5 min read

Agents

MCP explained: connecting AI to your tools safely

7 min read

Tutorials

RAG that actually retrieves: chunking, ranking and testing

12 min read

LLMOps

Designing an LLM gateway for a large organisation

9 min read