February 4, 2026
Pipes, Pythons, and pitchforks
Show HN: SymDerive – A functional, stateless symbolic math library
AI-ready math tool drops on HN — Python style police arrive
TLDR: SymDerive promises agent-friendly, step-by-step symbolic math with familiar Wolfram-style commands in Python. The comments explode over bad Python style and demand evidence that the “pipe” workflow helps AI, turning a neat launch into a “prove it” showdown that matters for anyone building reliable AI math tools.
A physicist-turned-quant just dropped SymDerive, a symbolic math library aimed at AI agents and humans who like tidy, step-by-step math. Think “no memory between steps” and a pipe you push expressions through until they come out simplified. It even speaks Wolfram syntax (hello, Integrate and Sin) to woo the Mathematica crowd, while running on Python under the hood. The repo is here: deriver.
But the community instantly turned into a courtroom drama. The loudest siren? Style cops: one commenter slammed the demo’s import * as “never ever,” calling it the programming equivalent of leaving your keys in the door. Another hit the “Pipe is better for agents” claim with a big “show me the receipts”, demanding proof it actually helps AI write code more reliably.
Jokes flew about the creator’s plea, “I will cry if roasted too hard”—the crowd obliged with a gentle roast and some memes about Python’s sacred rulebook. Fans liked the clean, stateless vibe and familiar math names; skeptics saw a flashy wrapper around existing tools and wanted benchmarks, not vibes. Bottom line: cool idea, spicy execution, and a classic Hacker News split—neat toy vs. prove it.
Key Points
- •SymDerive is a functional, stateless symbolic math library designed to be agent-native and useful for human users.
- •It enforces an Input → Transform → Output pipeline, favoring Lisp-style functional flows to improve reliability.
- •The library wraps SymPy, PySR, and CVXPY, while preserving Mathematica-like syntax to ease transition to Python.
- •Heavier capabilities (symbolic regression, convex optimization) are optional installs to avoid environment bloat.
- •Physics-oriented tools include abstract index notation for GR and Kramers-Kronig for causal models; orchestrators like Claude Code reportedly integrate well.