Mistral AI
Research Engineer, ML Platform
Palo Alto · Hybrid · Full-time · Posted Sep 3, 2026
- Experience
- 4+ years in ML infrastructure, distributed systems, Kubernetes platform engineering, or a related field.
About the job
Build and operate the ML platform and infrastructure for large-scale distributed training, evaluation, and batch inference across clusters and GPU hardware.
What you'll do
- Develop platform services, APIs, controllers, and tools for ML workloads, including scheduling and multi-cluster execution.
- Improve GPU capacity allocation, workload performance, reliability, observability, and self-service researcher workflows.
- Operate platform systems, including on-call participation and troubleshooting across applications, orchestration, networking, storage, and GPU infrastructure.
What they ask for
- 4+ years of experience in ML infrastructure, distributed systems, Kubernetes platform engineering, or a related field.
- Proficiency in Python or Go and comfort with production-grade distributed systems.
- Strong Kubernetes knowledge, including controllers, operators, CRDs, scheduling, networking, storage, and resource management.
- Understanding of distributed ML workloads and GPU infrastructure, including training, fine-tuning, evaluation, checkpointing, batch inference, and relevant technologies.
Nice to have
- Understanding technologies such as Kueue, Karpenter, Volcano, and Kyverno.
- Familiarity with PyTorch, CUDA, NCCL, and high-performance networking.
Tools
Python, Go, Kubernetes, Kueue, Karpenter, Volcano, Kyverno, PyTorch
AI focus
Build and operate infrastructure for large-scale ML training, evaluation, fine-tuning, and batch inference.
Model-written summary · the original posting is authoritative