Hi Andrew,
Thanks for sharing this interesting PR and your background in aerospace optimization — very relevant challenges (diverging iterations, strict domain constraints, need for backtracking).
In case you’re looking for alternatives/complements with full transparency (step-by-step logs, no black boxes), I recently released Opt4Deck: a small pure-Python optimization toolkit (Simplex, BFGS, Genetic, Adjoint) designed mainly for educational and prototyping use in engineering contexts. It might offer some ideas for constrained/iterative problems.
GitHub: GitHub - Opt4Deck/Opt4Deck: Opt4Deck is a unified optimization platform that integrates linear, nonlinear, evolutionary, and dynamic methods into a coherent and transparent environment. Designed for engineers and researchers, it provides clear methodology, computation transparency, and practical tools for reliably solving complex optimization problems. · GitHub
Site: https://opt4deck-teal-orca.reflex.run
Would love any quick thoughts if you have a moment!
Discussion: Opt4Deck: Transparent pure-Python optimization for education – SciPy feedback
Best,
Giannis Serafeim
#Opt4Deck