Hello HN! After several years as a data-scientist/engineer I was surprised by the lack of options for deploying mathematical optimization models, especially co

Hello HN! After several years as a data-scientist/engineer I was surprised by the lack of options for deploying mathematical optimization models, especially compared to the machine learning world. There are great libraries (Pyomo, JuMP, ...) but few end-to-end solutions to go from prototype idea to production API, and even fewer which provide strong mathematical consistency guarantees. This is where Opvious comes in... Its features include: + A declarative API to create models (linear, mixed-integer, quadratic) and automatically generate their LaTeX definitions. + Built-in productivity and debugging capabilities: multi-objective strategies, smart infeasibility detection, numerical performance insights… + A variety of integrations so you can solve problems from almost anywhere, for example pandas-compatible APIs for data pipelines and a TypeScript client to embed optimization directly in a web app. If you are interested in trying it out, the best place to get started is the welcome guide which walks through an interactive end-to-end example: https://www.opvious.io/notebooks/retro/notebooks/?path=guides/welcome.ipynb https://www.opvious.io/notebooks/retro/notebooks/?path=guide... The platform is free for non-commercial use, no separate solver installation or license required. I hope you find it useful and would love to hear your thoughts!