Search
Go
Member of Technical Staff - ML Infrastructure Engineer, Post-training
Summary by AIEngineer.careers
Member of Technical Staff - ML Infrastructure Engineer, Post-training is a ML Infrastructure Engineer role (contract). with Preference Model. in SAN FRANCISCO, US. Compensation shown: $200K–$350K. Imported listing (source: emploive.com). Apply on the employer's site (emploive.com).
Imported listing (source: emploive.com) · Apply on emploive.com
SAN FRANCISCO · US
$200K–$350K
Contract
SENIOR
The sections below reproduce the third-party job description for reference. AIEngineer.careers does not write or control this text.
Imported job description
Sourced from emploive.com
Preference Model is seeking a Senior ML Infrastructure Engineer to build scalable systems for post-training large language models. The role focuses on designing compute, scheduling, and data infrastructure for in-house RL environments, enabling high-throughput research and self-directed learning. You will partner with researchers to translate needs into robust engineering solutions.
Responsibilities
- Design and scale compute, scheduling, and data infrastructure for post-training LLM research
- Develop core ML framework primitives and internal tooling for reproducible experimentation
- Build evaluation, benchmarking, monitoring, and debugging tooling
- Partner with Research Engineers to implement infrastructure requirements rapidly
Requirements
- Strong software engineering fundamentals with experience building production-grade LLM training/inference infrastructure
- Experience with LLM internals (transformers, distributed training) and libraries like vLLM, SGLang, or Megatron
- Experience with RL training frameworks such as Slime, veRL, Ray Train, or SkyRL
- Deep understanding of distributed systems, cloud platforms (AWS/GCP), and Kubernetes
- Proficiency in PyTorch or JAX
Preference Model
San Francisco · US · 20+ employees
Preference Model builds reinforcement learning (RL) environments that automate machine learning (ML) research and engineering. The company develops infrastructure software to support RL experimentation and partners with frontier AI labs to advance LLM capabilities.
