Chai Discovery
Research Engineer, ML Infrastructure
San Francisco office · Full-time · Posted Sep 24, 2026
- Experience
- Deep industry experience working with top AI/ML teams; no years specified.
About the job
Develop core ML training and evaluation frameworks to make Chai’s models performant, resource-efficient, and reliable at scale.
What you'll do
- Build and optimize the training stack across model, layer, and kernel levels, including parallelism, quantization, and custom kernels.
- Profile large-GPU-cluster training runs, remove bottlenecks and failures, and monitor throughput, utilization, and uptime.
- Scale new architectures and training recipes while improving fault tolerance, checkpointing, and deterministic orchestration.
What they ask for
- Strong software system design skills.
- Proficiency in Python.
- Fluency in Pytorch or JAX.
- Demonstrated exceptional impact on real-world problems and systems.
Tools
Python, Pytorch, JAX
AI focus
Build and optimize training and evaluation infrastructure for frontier models used in AI drug design and therapeutic engineering.
Model-written summary · the original posting is authoritative