Hi Ayush,
I came across your profile — your experience building and evaluating machine learning models at Amazon, improving model performance, running experiments, and working with production-scale AWS infrastructure really caught my attention.
I'm a software engineer in Japan, and I built FaultWright myself: it deliberately breaks software in realistic ways, has an AI try to diagnose and fix the break, then checks whether the fix actually worked. The engineering side is already handled. What I'm looking for now is someone who can help connect the system to real AI and engineering needs — identifying meaningful evaluation scenarios, understanding what model teams actually care about, talking with potential users, and helping turn this into paid AI-training and evaluation work.
Your experience evaluating models with metrics like precision and recall, running performance experiments, and improving ML systems seems especially relevant to the kind of technically strong collaborator I'm looking for.
To start, this would be a flexible, revenue-share collaboration rather than a salaried role. If we work well together and our goals align, I'd be glad to discuss a longer-term partnership, including a potential co-founder role.
Happy to walk you through what I've built if you're interested — let me know.