AI systems that survive production
Most AI agents in production are demos with extra steps. I help teams build the ones that aren't.
I'm Bhanu Chaddha. I work with teams putting AI agents and RAG into production: the parts that hold up once the demo is over.
- 48
- posts published
- 10 / 21
- Agents in Production
- 10 / 10
- Docker, Step by Step
- 1 / 12
- The Delivery Layer
- 1
- The AI Agent Playbook
Fifteen years building distributed systems before AI is why the production critique lands. About me
What I write about
All topics →Recent posts
View all →A Prompt Edit Is a Deploy. Treat It Like One.
Agents in Production · part 10
A prompt edit changes what your agent does for every user, just like code. Give it review, tests, staged rollout, and a rollback button, before it needs one.
The Decision You Didn't Make
The Delivery Layer · part 1
Most project damage comes from choices nobody recorded making. A bad call can be reviewed and reversed. An unmade decision just accrues until it breaks.
Model Routing Is an Optimisation, Not an Architecture
Agents in Production · part 9
Routing is a cost optimisation, not day-one architecture. Build it once you have production data showing which requests a cheaper model actually handles well.
Your Tool Schema Is a Prompt, Not an API
Agents in Production · part 8
A tool schema is read by the model on every call, not once by an engineer. Design the contract, the errors, and the catalog size, because MCP will not.
Working on something in this space? I'm happy to talk it through.