Your DevOps investigations shouldn't die in a closed browser tab.
Every time an DevOps/Platform engineer/SRE debugs a CrashLoopBackOff or traces why an EC2 instance got stopped, that hard-won context usually vanishes the moment the incident closes. The next person (or the next you, 3 months later) starts from zero.
So we shipped memory into DevOps Open Agent.
Introducing RAG powered by Qdrant ðŸ§
Here's what it does: → Every completed investigation - Kubernetes and AWS - is embedded and stored as a vector in Qdrant → On your next investigation, tick "Include past investigations (RAG)" → The agent retrieves the most similar prior cases and feeds them to the LLM as context → Root cause analysis now builds on what your team has actually seen and fixed before
It turns a pile of one-off investigations into institutional memory - recurring root causes, the fixes that worked, the dead ends you already ruled out.
A few things I'm proud of:
✅ Bring your own Qdrant (local Docker or Qdrant Cloud) - you own your data
✅ Embeddings reuse your existing LLM provider (OpenAI, Gemini, or Ollama) - no new keys required
✅ Retrieval is scoped to your own history
✅ 100% open source and self-hostable
Reactive → proactive → now with memory.
The vision: DevOps troubleshooting that gets smarter with every incident, not one that forgets.
Configure under Integrations → Qdrant(RAG):
🔗 github.com/ideaweaver-ai/devops-open-agent
🎥 Demo: https://lnkd.in/eBWcPRtj
🎓 Want Live Instructor-Led Training?
If you'd like hands-on, instructor-led sessions where you can ask questions and get your doubts resolved live, check out our courses:
Top comments (0)