Hi HN, I've been building MolBuilder -- a pure-Python molecular engineering toolkit that covers the full pipeline from molecular structure through retrosynthesis, reactor selection, safety assessment, cost estimation, and scale-up analysis.

Latest feature: substrate-aware condition prediction. It analyzes steric/electronic properties of your molecule and ranks compatible reaction templates with optimized conditions.

The install story is the differentiator: pip install molbuilder vs. RDKit's conda/C++ compilation. It's not trying to replace RDKit's computational accuracy -- it's solving a different problem: going from "I have a molecule" to "here's how to make it at scale."

1,280+ tests, Apache 2.0 license, Python 3.11+. Also available as a REST API.

  - PyPI: pip install molbuilder
  - Tutorials: https://github.com/Taylor-C-Powell/Molecule_Builder/tree/main/tutorials

Happy to answer questions about the implementation or design choices.