Staff Engineer, Research Infrastructure & ML Platform

Simile; San Francisco (CA); On-site; USD 200,000 - 400,000; Full time

Benefits offered by this job

Equity grants

Health, dental, vision

Flexible time off

Job summary

Simile, the simulation company in San Francisco, seeks a Member of Technical Staff in Research Infrastructure to build and own the ML platform researchers rely on—from data schemas to production services serving millions of interdependent agent calls.

You will optimize training and inference pipelines, lead data architecture for scale, and own the GPU cluster. Ideal candidates have deep ML/system experience, Python, PyTorch/JAX, and a track record shipping production ML platforms.

Qualifications

• Strong systems and ML proficiency with Python.
• Experience with modern ML frameworks (PyTorch/JAX).
• Production ML platforms or MLOps experience.
• Experience architecting and debugging production distributed systems.
• Own deployment pipelines from data to serving.
• Research- and data-literate with ability to navigate ML frontier.
• Self-directed and pragmatic with team collaboration.
• Strong quantitative foundation in CS/Math/Stats.

Responsibilities

• Build the ML platform our researchers rely on, from data schemas to production serving.
• Make training and data pipelines fast, profiling GPU memory and FLOPs.
• Make serving fast and cost-effective for population-scale simulations.
• Lead data-architecture redesign to handle high-volume simulations.
• Own the multi-node GPU cluster health, scaling, and autoscaling.
• Engineer evaluation tooling with rigorous statistical approaches.
• Push state of the art by translating academic work into production.

Skills

Python

PyTorch

JAX

Distributed systems

MLOps

GPU

Research & data literacy

Self-directed

Teamwork

Education

B.S. in Computer Science or related field

Tools

NVIDIA GPUs

CUDA

NCCL

InfiniBand

NVLink

Triton

TensorRT-LLM

vLLM

Job description

Simile, the simulation company in San Francisco, seeks a Member of Technical Staff in Research Infrastructure to build and own the ML platform researchers rely on—from data schemas to production services serving millions of interdependent agent calls.

You will optimize training and inference pipelines, lead data architecture for scale, and own the GPU cluster. Ideal candidates have deep ML/system experience, Python, PyTorch/JAX, and a track record shipping production ML platforms.