Senior ML Infra Engineer — LLM Training & RL Research

Preference Model; San Francisco (CA); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

Competitive cash and equity

Ownership and autonomy

Visa sponsorship

Relocation support

Lunch onsite

Weekly snack orders

Job summary

Preference Model is a pioneering AI research company building RL environments for training capable and well-aligned superintelligences. We are seeking Senior ML Infrastructure Engineers to design and scale the infrastructure powering post-training research on large language models.

You will collaborate with researchers to enable high-throughput experiments and reliable training workflows. Join a fast-moving startup backed by significant seed funding, with opportunities to work alongside senior

Qualifications

• Strong software engineering fundamentals.
• Experience building production-grade LLM inference and training infrastructure.
• Experience with distributed training and ML frameworks like PyTorch or JAX.
• Experience with cloud platforms (AWS, GCP) and container orchestration (Kubernetes).
• Experience building robust data pipelines and infrastructure tooling.
• Ability to translate research needs into scalable infrastructure.

Responsibilities

• Design, build, and scale compute, scheduling, and data infrastructure for post-training research.
• Develop and maintain ML framework primitives and internal tooling for reproducible experiments.
• Build evaluation, monitoring, logging, debugging, and automated testing/deployment systems.
• Collaborate with Research Engineers to translate research needs into infrastructure requirements.

Skills

LLM infrastructure

Distributed systems

PyTorch

Kubernetes

Cloud platforms

Tools

Kubernetes

vLLM

Megatron

Job description

Preference Model is a pioneering AI research company building RL environments for training capable and well-aligned superintelligences. We are seeking Senior ML Infrastructure Engineers to design and scale the infrastructure powering post-training research on large language models.

You will collaborate with researchers to enable high-throughput experiments and reliable training workflows. Join a fast-moving startup backed by significant seed funding, with opportunities to work alongside senior