ML Systems Engineer: Post-Training Pipelines & RL

ThirdLayer, Inc.; San Francisco (CA); On-site; USD 140,000 - 210,000; Full time

Job summary

ThirdLayer, Inc. is seeking a Research Engineer to build end-to-end training pipelines and infrastructure that support continuous post-training runs. You will bridge research and production, turning experiments into scalable, reliable systems that run with production data.

You will work with researchers to co-design methods and systems and partner with product engineers to ship impactful improvements across our RL and data platforms.

Qualifications

• Strong software engineering fundamentals with ML depth.
• Experience with distributed training or large-scale experiments.
• Proficiency in Python and ML frameworks (PyTorch, JAX).
• Ability to turn prototyping into production-ready systems.
• Experience with RL environments or data quality systems is a plus.

Responsibilities

• Build and own pipelines for post-training runs from data ingestion to deployment.
• Turn research prototypes into reliable systems that run on production data.
• Create infrastructure to generate, scale, and version training environments and evals.
• Develop tooling to extract signals from traces and curate data for learning loops.
• Build instrumentation to inspect, debug, and understand training runs.
• Collaborate with researchers and product engineers to ship results.

Skills

Distributed training

ML systems

Python

PyTorch

JAX

Data pipelines

System design

Job description

ThirdLayer, Inc. is seeking a Research Engineer to build end-to-end training pipelines and infrastructure that support continuous post-training runs. You will bridge research and production, turning experiments into scalable, reliable systems that run with production data.

You will work with researchers to co-design methods and systems and partner with product engineers to ship impactful improvements across our RL and data platforms.