Diktiax: a Rust/Next.js agent that hunts insurance fraud rings via graph memory
JeremyCMorgan · x · 2026-11-19
- Diktiax is an open-source insurance fraud investigation agent (Rust/Axum backend, Next.js frontend) that finds fraud rings by walking a FalkorDB graph.
- Key architecture: graph evidence and deterministic scoring are separated from LLM decisions; every verdict joins a constraint-guarded chain in the graph that the agent — and any other agent via MCP — reads before investigating again, effectively graph-based agent memory.
- Built for FalkorDB's 'Graph Hacks' hackathon Tracks 01/02; ships with sample data, Docker setup, tests, benchmarks, and a live demo.
Related event: Diktiax: Open-Source Graph AI Agent for Insurance Fraud Investigation(2 posts)→
More from coding & agent
- Devs flag GPT's defensive programming for hiding bugs; OpenAI asks for /feedback cases — pvncher · 2026-11-19
- Flutter ships A2UI 1.0: agents respond with real interactive widgets, not chat text — rseroter · 2026-11-19
- Skills vs MCP vs RAG vs Memory: the four context mechanisms AI agents actually need — MaryamMiradi · 2026-11-19
- Anthropic details agent memory: 'dreaming' lets Claude organize its experience — daniel_mac8 · 2026-11-19
- The case for externalized memory as the real fix for continual learning — daniel_mac8 · 2026-11-19
- FT: US tech outages blamed on rush to fix bugs found by AI — nordicinst · 2026-11-19