Runlayer Labs
Notes from the team building Runlayer Agents. We write up what we ship, how it works under the hood, and what changed when we ran it on our own work.
All posts
6 posts-
Manage Runlayer Agents as code with Terraform
Runlayer Agents can now be defined in Terraform: the prompt, model, tools, and network access live in your repo, and every change ships through a reviewed plan.
-
Zero-trust networking for Runlayer Agents
An agent that can reach any host on the internet can leak anything it can read. Runlayer Agents now carry their own network access policy, checked on every connection the agent makes.
-
Runlayer Agents are now in Microsoft Teams
Runlayer Agents now work inside Microsoft Teams. Each agent gets its own bot and keeps the identity, policies, and audit trail it has everywhere else in Runlayer, with nothing for your team to set up in Azure.
-
Build your own software factory with Runlayer Agents
Ramp built Inspect. Stripe built Minions. The strongest engineering teams are building factories that build their software. Here’s how we built ours on Runlayer Agents, and how you can build yours without a dedicated platform team.
-
What did your AI spend actually produce?
Most companies can tell you what they spend on AI. Very few can tell you what that spend produced. Runlayer now connects AI spend to the real work it created, by team, department, and agent.