Building a persistent customer memory layer for a support agent with Hindsight
Front_Xordinary0905 · reddit · 2026-09-29
The author walks through engineering a persistent memory layer for a customer-support agent so it stops re-asking returning customers for information they already gave. Key points: a two-layer design separating MemoryService (app-level memory interface, per-customer memory banks, local fallback dict for when Hindsight Cloud is unconfigured/failing) from HindsightClient (auth, HTTP, endpoints, response parsing); a flow of Customer → Agent → MemoryService → HindsightClient → Hindsight Cloud → retrieved context; and storing both sides of each interaction as one combined memory block ("User: …\nAgent: …") so future retrieval carries full conversational context, illustrated with a recurring Wi-Fi outage support scenario.
More from coding & agent
- Claude Opus recreates the magical sheet-music scene from a screenshot and a free wav — justin_hart · 2026-09-29
- OpenAI launches always-on Workspace Agents in ChatGPT with Slack deploys and MCP support — testingcatalog · 2026-09-29
- BaRe-Mem: Bayesian Reliability Memory Makes Multi-Agent Consultation Robust to Misleading Advisors — NanyangTechnologicalUniversity · 2026-09-29
- SkillDRE Evolves Malicious Agent Skills via Dual-Stage Feedback, 45.28% Attack Success — Pengyu Zhu · 2026-09-29
- 19-year-old claims $750K profit from Claude Code arbitrage bot built in 2 days — Aiden_Tech_Ai · 2026-09-29
- Agent edits the workflow at build time, plain script at run time: token-saving browser automation — ZennoLab_Guru · 2026-09-29