Shmungus

Your niche is too specific for the analyst firms and too small to staff a research team. Tensorust finds the exact public record your question lives in, connects it, and hands you the answer your larger competitor pays…

I've spent a few years building quant strategies and most of the work isn't the math — it's the plumbing: getting clean data, structuring the backtest so it doesn't look at the future, generating a readable tearsheet. I…

Proof Engine renders mathematics, not graphics. A Lorenz attractor looks like a Lorenz attractor because particles follow the actual differential equations in real time. Entities are held together by force fields and…

I couldn't find a Rust crate that gave me validated financial types backed by decimal arithmetic. Everything I found either used f64 (unacceptable for order books), panicked on bad input, or was a thin wrapper around a…

I've been building the substrate layer for production AI systems in Rust. This is the index repo for 10 crates I shipped this week. The AI side covers agent memory (episodic, semantic, working with decay and multi-agent…

- llm-stream — streaming from OpenAI + Anthropic, callback-based - llm-cache — file-backed semantic cache, LRU eviction - llm-cost — offline token counting + cost estimation - llm-retry — exponential backoff + circuit…

I built this after getting frustrated with "multi-agent" frameworks that claim parallelism but are really just one fat async task with no resource bounds. tokio-prompt-orchestrator breaks LLM inference into 5 physical…