ML Performance Engineering Manager

Polluxa, Inc.; Kirkland (WA); On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Health insurance

Dental insurance

Vision insurance

Life insurance

Disability insurance

401(k) with company match

Paid time off

Sick time

Maternity leave

Baby bonding leave

Holidays

Job summary

Polluxa, Inc. is seeking a senior ML leader in Kirkland, WA to guide a team of engineers focused on ML training and serving benchmarks representative of industry needs.

This role emphasizes performance optimization for launch readiness and OSS benchmark submissions, with emphasis on TPU features and model-preserving techniques. You will collaborate with cross-functional partners to push the envelope on hardware-aware ML pipelines, automate benchmarking, and mentor a growing team of researchers

Qualifications

• Bachelor’s degree or equivalent practical experience.
• 8+ years of experience in software development.
• 5 years of experience leading ML design and optimizing ML infrastructure.
• 3 years in a technical leadership role.
• 2 years in people management or team leadership.
• Experience with ML performance analysis, benchmarking, and computer architecture.

Responsibilities

• Lead a team of software engineers focused on ML training and serving benchmarks representing Google production and the broader ML industry.
• Achieve performance for customer launches and, for OSS models, for benchmark submissions (ML Commons, InferenceX, etc.).
• Use benchmarks to identify performance opportunities and drive compiler/runtime optimizations and related tooling.
• Participate in algorithmic innovations exploiting TPU hardware features and model-preserving optimizations.
• Collaborate on co-designing TPU-friendly models to showcase quality against GPU-centric OSS models.

Skills

ML leadership

Benchmarking

Performance analysis

System optimization

Team management

CUDA

TPU knowledge

PyTorch

JAX

Education

Bachelor's degree or equivalent

Master's degree or PhD preferred

Tools

CUDA

Triton

Pallas

MLIR/OpenXLA

PyTorch

JAX

vLLM

Job description

Polluxa, Inc. is seeking a senior ML leader in Kirkland, WA to guide a team of engineers focused on ML training and serving benchmarks representative of industry needs.

This role emphasizes performance optimization for launch readiness and OSS benchmark submissions, with emphasis on TPU features and model-preserving techniques. You will collaborate with cross-functional partners to push the envelope on hardware-aware ML pipelines, automate benchmarking, and mentor a growing team of researchers