ML Systems Engineer — Training & Inference Infrastructure

NextToken; Mountain View (CA); On-site; USD 190,000 - 240,000; Full time

Job summary

NextToken is seeking engineers to build robust ML infrastructure that enables model improvement through scalable data generation, training pipelines, and deployment tools. You’ll empower researchers and product teams to run experiments and deploy checkpoints without rebuilding workflows.

You’ll work across data pipelines, distributed execution, training, and serving with a focus on reliability and performance.

Qualifications

• Experience building ML infrastructure for training systems or model serving.
• Ability to profile workloads and diagnose performance issues in ML pipelines.
• Strong distributed systems fundamentals: scheduling, storage, and fault tolerance.

Responsibilities

• Develop data generation and experiment workflows to enable reliable model improvement.
• Build scalable training and evaluation pipelines across distributed environments.
• Improve inference performance, including batching, memory management, and deployment considerations.
• Create benchmarks and tooling to support researchers and customers.

Skills

ML infrastructure

Distributed systems

PyTorch/JAX knowledge

Tools

CUDA

Triton

Container orchestration

Job description

NextToken is seeking engineers to build robust ML infrastructure that enables model improvement through scalable data generation, training pipelines, and deployment tools. You’ll empower researchers and product teams to run experiments and deploy checkpoints without rebuilding workflows.

You’ll work across data pipelines, distributed execution, training, and serving with a focus on reliability and performance.