Research Engineer, ML Infrastructure

The role

You will develop core frameworks for model training and evaluation to make models performant, resource efficient and reliable at scale. This involves building the training stack and profiling training runs on large GPU clusters.

What you will do

• Build the training stack across model, layer, and kernel levels
• Optimize workloads through parallelism, quantization, and custom kernels
• Profile end-to-end training runs on large GPU clusters
• Eliminate bottlenecks and failures
• Monitor throughput, utilization, and uptime
• Ensure new model architectures and training recipes scale efficiently
• Make the ML training stack maximally reliable through fault tolerance, checkpointing, and deterministic orchestration

Required qualifications

• Strong software system design skills
• Proficiency in Python
• Pytorch or JAX fluency

Nice to have

• Deep industry experience working with top AI/ML teams on the kinds of problems and systems described

About the company

Chai Discovery builds the computer-aided design suite for molecules. Companies like Eli Lilly, GSK, Pfizer, Bristol Myers Squibb, argenx and Novartis use Chai to power drug discovery programs.