Terraform for agents: how we define our software factory in code
All the agents at our company run out this one configuration file: It’s kind of like… Terraform for agents. It started as a place to save some configuration for cloud runners, like Docker container
All the agents at our company run out this one configuration file: It’s kind of like… Terraform for agents. It started as a place to save some configuration for cloud runners, like Docker container information and repositories to clone. Since then, just about everything our agents do is configured as code: Agent orchestration patterns and model selections Which secrets and MCP servers agents can access Automations and triggers that kick off cloud agents Strategies for agent scoring, benchmarking
more from @BHolmesDev
So, you want to build a software factory?
More and more engineers seem to be investigating software factories. There are many ways to define “software factory”, but here’s the one we’ve settled on: centralizing your system of agents into one
The road to self-improving software
We’re all starting to think in systems to get better code out of agents. Maybe you’ve explored adding loops for computer use verification, auto-fixing CI failures, and adversarial code review to get