I've been a heavy user of coding agents for over a year now, and while excited by the productivity gains - I've increasingly been frustrated at the process around it.
- Prompts are imperative and ephemeral, and code is increasingly a black-box, so I lose track of intent over long horizons.
- Pull requests, CI/CD, and even GitHub have become process bottlenecks and are increasingly losing meaning in a world of code abundance.
- I'm tired of babysitting coding agents, cleaning up slop, and reviewing AI output!
I've been building Telos to solve some of these problems - it's a new, declarative abstraction layer for building software with AI. We treat the goal specification as the source of truth that AI coding agents continuously reconcile against in a cloud environment.
We've been dogfooding this quietly for a few months, and it's solved many of the problems above for us - we're excited to share this with the world!
Our agent runtime's source is available [0], and instructions to try it out alongside a real-world example are at the end of the linked post.
Cheers