Redditor proposes graph-based deterministic modeling to make LLM finance agents trustworthy
jonnylegs · reddit · 2026-10-07
Reddit user jonnylegs argues LLM-based finance agents fail because models don't truly compute: simple arithmetic is pattern recall, mortgage math relies on ad-hoc codegen with no determinism guarantee, and for complex multi-period models the construction steps get lost, leaving users unable to validate results.
His solution: a node-based graph system he self-funded years ago for modeling financial calculations — previously blocked by a steep learning curve, but now LLMs can understand the concepts and build models on top of it. He's in proof-of-concept with a lemonade-stand demo video and plans longer deep dives to gather feedback.
More from coding & agent
- gpt-live-1 called a generation ahead for voice AI assistants: duplex + delegation — pbbakkum · 2026-10-07
- Blenderbox: open-source multiplexer giving coding agents isolated headless Blender sessions — tobowers · 2026-10-07
- Models aren't the bottleneck: report maps 4 levels of agentic software development — rseroter · 2026-10-07
- Fireworks launches Nexus: smart routing to open models cuts coding AI bills in half with 95%+ cache hits — sophiamyang · 2026-10-07
- YC's Paxel analyzed 3.4M AI coding sessions — and privacy questions follow — AaronBergman18 · 2026-10-07
- EmbeddingGemma can already self-host into OpenClaw memory via Ollama or OpenAI-compatible APIs — osanseviero · 2026-10-07