The PyTorch compiler team is the driving force behind PT2, a step function change in the history of PyTorch. PT2 technologies have gained industry-wide recognition since its first OSS release in 2023. But there is still a long and exciting technical roadmap going forward. We seek an Engineer Manager to drive subcomponents of the PyTorch Compiler development and holistic areas across teams, such as vertical performance, heterogeneous hardware, PT2 wider adoption, and compiler-accelerated large-scale distributed training. The team is highly innovative and passionate about the technologies we build. The work is cutting-edge and industry-influencing, so we look for leaders with technical expertise who demonstrate strong initiatives (to push through the cutting-edge) and exceptional productivity (to keep up with the team's pace).
Software Engineering Manager, PyTorch Compiler Responsibilities

• Support a team of highly passionate and highly innovative compiler specialists in a cutting-edge domain.
• Lead with a strong sense of direction and be well-informed of industry trends about or adjacent to our space.
• Bootstrap new initiatives, especially across teams (e.g., defining what success looks like, creating project narrative, getting buy-in, building missing structure).
• Drive to committed business outcomes, know what is happening, and have a healthy information channel.
• Be what the team needs us to be
• wear different hats (EM, TPM, PM, strategist, planner, communicator, reporting STO) as necessary.

Minimum Qualifications

• Experience with PyTorch, ML runtime, or ML compilers.
• 2 years of experience managing a team of ML framework, runtime, or compiler engineers.
• Leadership experience by setting directions, driving strategies, and establishing cross-functional collaborations.
• Demonstrated experience in recruiting, structuring, and leading technical organizations, including performance management.

Preferred Qualifications

• Knowledge of or experience with PT2 technologies.
• Experience with other ML compilers such as XLA, TVM, ONNX, Halide, and Triton.
• Experience in vertical performance optimization, GPU performance, quantization, or distributed training.
• Knowledge of ML model architecture research landscape

Start preparing
Learn about how to prepare for your interview with our interview guide, tips, and interactive experiences.
Visit interview prep