3 months ago
Responsibilities
- Develop high-quality open-source software for Ray to simplify distributed programming.
- Identify, implement, and evaluate architectural improvements to Ray Core.
- Lead cross-team projects and mentor junior team members.
- Improve Ray’s testing process and release stability.
- Work on performance optimization, stability and stress-testing infrastructure, and fault tolerance.
- Communicate technical work through talks, tutorials, and blog posts.
Requirements
- At least five years of relevant work experience is required.
- Experience building scalable and fault-tolerant distributed systems is required.
- Extensive experience with C/C++ and low-level operating systems is required.
- A solid background in algorithms, data structures, and system design is required.
- Knowledge of distributed model training and inference, including tensor parallelism and pipeline parallelism, is preferred.
- Knowledge of GPU programming is preferred.
About Anyscale
Anyscale enables Python developers to build and run all their AI—from data prep to training and inference—at any scale. Anyscale is trusted by leading AI teams at Canva, TripAdvisor, Physical Intelligence, Coinbase and more.