Jason Liu is an MLE with 11 years of experience in NYC who bridges research and production across LLMs, computer vision, and recommendation systems. At Stitch Fix he built a 300k-outfit dataset, led CV research and prompt engineering efforts, and engineered a vector product-similarity search (CLIP + GPT-3 embeddings + FAISS) that contributed over $50M in annual revenue while also architecting Flight, a recommendation framework serving 350M requests/day with >80% adoption. His technical strengths include retrieval-augmented generation, query understanding, embedding fine-tuning, prompt engineering, and MLOps observability, and he contributes backend work to open-source LLM tooling such as instructor-ai/instructor. Now advising startups through 567 Labs and heads-down at South Park Commons, he helps teams prioritize experiments, vet talent, and simplify tooling to get reliable ML products into production. He holds a Bachelor's in Computational and Applied Mathematics from the University of Waterloo.
Contributions:62 releases, 450 reviews, 691 PRs in 1 year 9 months
Contributions summary:Jason appears to be a back-end developer, with the primary focus on creating and modifying code related to OpenAI's functions. Their contributions include creating and modifying core functionalities for building and invoking functions, especially the `openai_function` and `OpenAISchema` classes and how to handle results from responses. They also implemented a working segmentation example.
Contributions:1 review, 27 commits, 7 PRs in 1 month
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.