Zeina Migeed
Research Scientist at Meta
San Francisco Bay Area United States
Summary
🤩
Rockstar🎓
Top SchoolZeina Migeed is a research scientist and PhD candidate in Programming Languages at UCLA with 11 years of experience bridging PL theory and practical engineering. She interned and now works at Meta, where she applied gradual typing and SMT-backed shape reasoning to PyTorch, contributing merged code to the flagship pytorch/pytorch repo and shipping tools that remove control flow and catch shape errors across real workloads. Her work combines formal results (POPL publication) with substantial engineering—6k LOC across frameworks and integrations with Z3—to make mixed-typed Python code safer and more analyzable. Zeina’s research introduced the migration space concept and novel inference techniques that handle symbolic relations and algebraic expressions, improving both developer tooling and documentation. Based in the Bay Area, she designs sound, practical type systems for industry-relevant languages and is exploring broader applications of static analysis to IDEs, security, and correctness guarantees. An underrated strength is her ability to deliver production-grade implementations of cutting-edge PL research that scale to large ML codebases.
11 years of coding experience
5 years of employment as a software developer
BS Computer Science and Mathematics Mathematics and Computer Science, BS Computer Science and Mathematics Mathematics and Computer Science at Northeastern UniversityUniversity of California, Los AngelesBS Computer Science Computer Science , BS Computer Science Computer Science at The American University in Cairo
English, Arabic, Hebrew
Github Skills (10)
pytorch10
python10
z310
type-checking9
typehinting9
tensor9
deep-learning8
neural-network8
autograd7
machine-learning7
Programming languages (4)
Github contributions (5)
pytorch/pytorch
Mar 2021 - Aug 2022
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
ML Engineer
ML EngineerContributions:52 reviews, 342 commits, 88 PRs in 1 year 5 months
Contributions summary:Zeina contributed to the development of gradual typing constraints for the PyTorch framework. Their commits focused on defining high-level type constraint definitions for FX graphs, particularly within the context of the experimental migrate_gradual_types module. The user implemented constraint generation rules for various operations, including addition, reshaping, and linear layers, and also integrated with the Z3 solver. These changes are designed to enhance the type safety and analysis capabilities of the PyTorch framework.
pythongpu-accelerationdeep-learninggpunumpy
migeed-z/Lisp_interpreter
May 2015 - Jul 2016
Contributions:93 pushes, 2 branches in 1 year 2 months
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.