This week our focus on DevOps Open Agent is still enterprise readiness, this time on the AI side.
A lot of teams already use AWS. They have IAM, regions, and Bedrock model access set up. Asking them to also manage a separate OpenAI or Anthropic key just to run investigations adds friction they don’t need.
That’s why I’m happy to share: DevOps Open Agent now supports AWS Bedrock as a shared LLM provider.
How it works
✅ Set LLM_PROVIDER=bedrock in backend/.env
✅ Pick your model (BEDROCK_MODEL) Claude, Llama, and more via the Converse API
✅ Uses the same AWS credential chain you already trust (~/.aws / AWS_PROFILE / IAM role)
✅ Optional region and profile overrides (BEDROCK_REGION, BEDROCK_AWS_PROFILE)
✅ Works across agents: Kubernetes, AWS, cost, PR review, and more, with the same shared LLM layer
So you can run AI root-cause analysis on Bedrock without leaving your AWS footprint.
Open source, self-hostable, built for DevOps engineers who already live in AWS.
If you like the product, please don’t forget to give us a ⭐ on GitHub.
If you’re running Bedrock in production for DevOps AI, I’d love your feedback on which models and guardrails you want next.
🔗 GitHub: https://lnkd.in/giri6MF5
🎥 YouTube: https://lnkd.in/g2PbG3Hc
Top comments (0)