AWS case study: DSPy RLM mines years of Slack history into an AI IT help desk knowledge base
lateinteraction · x · 2026-11-09
An AWS post on ShipStation Global's AI IT help desk (Bedrock AgentCore) spotlights a dspy.RLM module that treats the company's entire Slack history as a navigable memory space inside a sandboxed Python REPL. The three-stage pipeline: export full channel history via the Slack API into structured dumps; have the RLM iteratively classify each thread as resolved/unresolved/escalated with strict quality criteria (only threads containing a complete, workable fix count); then orchestrate recursive sub-LLM calls to extract structured fields validated against Pydantic models. Key insight: indexing unverified fixes makes the knowledge base surface wrong answers with the same confidence as correct ones.
More from coding & agent
- Swarm agent training needs thousands of agents, ambitious problems and cost math — samiramanabi · 2026-11-09
- product-engineering: MIT-licensed agent skills for shipping products with Claude agents — PawelHuryn · 2026-11-09
- MIT-licensed skill embeds Claude agents that can propose but not execute Stripe changes — PawelHuryn · 2026-11-09
- Google's frontier AI product head Tulsee Doshi on why PMs are suddenly valuable again — lennysan · 2026-11-09
- GitHub Copilot app adds Projects: give it a goal and it coordinates cross-repo work — 0xkarasy · 2026-11-09
- GitManual turns any GitHub repo into a step-by-step setup guide — adnan_hashmi · 2026-11-09