About

Hi, I'm Venkat.

I'm a product manager in the Bay Area working on AI agents for regulated enterprises.

At home I build and run my own: Miya, my personal AI agent, and Rahat, the platform underneath it. Seven specialist agents share one memory, one rulebook and one record of every decision, and they run in my family's life every day, which means I feel every failure the same day.

I care about agents you'd actually point at your own life: ones that remember, coordinate, check themselves, and know when not to act. Most of what goes wrong with agents isn't in the model. It's in the layer underneath, and that's the layer I like to build.

What's here

• Rahat: the system, what's running and what isn't, and the code.
• Writing: what it takes to trust an agent, starting with the series Where do we eat?
• Demos: something that shipped, every two weeks.

Get in touch

The best way to reach me is LinkedIn. The code is on GitHub, and the writing is on Substack. Happy to walk through any of it.

Views my own, not my employer's.