Sami Jaghouar
Founding Research Enginer at Prime intellect
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolSami Jaghouar is a research-focused software engineer with 6 years of experience building and scaling ML systems, currently founding research engineer at Prime Intellect in San Francisco, where he works on distributed large-model training in low-bandwidth, heterogeneous, and unreliable compute environments. He has hands-on experience training language models at scale (nyonic) and shipping production-ready ML infrastructure at Jina AI, contributing notable backend fixes and feature work to popular open-source projects like jina-ai/serve and docarray. His contributions reveal deep familiarity with request handling, executor parameters, multimodal data serialization, and memory backends—skills that bridge research prototypes and cloud-native deployments. Sami’s background includes published academic work in active search for traffic-sign recognition, reflecting a blend of rigorous research and pragmatic engineering. He pairs a strong theoretical foundation from UTC and Chalmers with a track record of quietly fixing hard edge cases that keep distributed ML systems reliable in production.
6 years of coding experience
4 years of employment as a software developer
Lycée ChaptalUnivsersity of Technology of Compiègne Mathematics and computer science, Univsersity of Technology of Compiègne Mathematics and computer science at Université de Technologie de Compiègne (UTC)Master exchange program Computer science and AI, Master exchange program Computer science and AI at Chalmers University of Technology
Github Skills (28)
fastapi10
python10
machine-learning10
numpy10
tensor10
data-structure9
pytorch9
protobuffer9
protobuf9
grpc9
data-structures9
microservices8
orchestra8
orchestration8
docker8
Programming languages (13)
Github contributions (5)
docarray/docarray
Jan 2022 - Jan 2023
Represent, send, store and search multimodal data
Role in this project:
Back-end Developer
Back-end DeveloperContributions:981 reviews, 137 commits, 536 PRs in 1 year
Contributions summary:Sami's commits primarily focus on enhancing the `docarray/docarray` repository. They addressed bugs related to deleting documents by ID and slice, implemented improvements within the memory backend, and made significant code changes to facilitate the integration of a filtering mechanism for matching methods. Further, the user refactored the stacking process and incorporated the handling of various tensor types with their respective proto serialization, along with fixing Fast API document response issues.
protobuffeature-storevectorunstructured-dataqdrant
jina-ai/serve
Jan 2022 - Jan 2023
☁️ Build multimodal AI applications with cloud-native stack
Role in this project:Backend Developer
Backend DeveloperContributions:300 reviews, 189 commits, 231 PRs in 1 year
Contributions summary:Sami primarily contributed to backend development, with a focus on fixing bugs and implementing new features related to the `peapods` and CLI components. They fixed naming issues and statement errors. The user also worked on Executor parameters implementation and modifications to the request handling. This indicates a strong understanding of the core workings and internal architecture of the project.
opentelemetryllm-inferencecreative-aigrpcproduction
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.