Latest AI JobsBrowse Jobs
Mistral AI logo

Research Engineer, ML Platform

Mistral AIPalo AltoScience
Apply on company site
Salary
Not listed
Job type
Full-time
Work mode
Remote
Posted
3 Sept 2026
Status
Open · checked 7d ago

About this role

Mistral AI is seeking a Research Engineer, ML Platform for a full-time, remote-friendly position based in Palo Alto. This role suits an experienced infrastructure specialist who wants to build and operate the foundational systems that power large-scale artificial intelligence training and inference.

Role Overview

The Research Engineer, ML Platform will focus on developing the infrastructure that allows researchers and engineers to execute distributed GPU workloads dependably across multiple clusters, hardware configurations, and geographic regions. The professional in this position handles the entire machine learning lifecycle, ranging from workload scheduling and resource capacity management to platform APIs, system observability, and production operations. Candidates take full ownership of critical systems to transform complex infrastructure into streamlined, self-service capabilities.

Key Responsibilities

Build comprehensive services, APIs, controllers, and tooling specifically for model training, evaluation, fine-tuning, and batch inference.
Create orchestration systems for workload queueing, admission control, quota management, priorities, preemption, and topology-aware placement.
Improve the provisioning, allocation, and utilization of heterogeneous GPU resources across diverse clusters.
Enable multi-cluster execution by routing workloads based on available capacity, data locality, hardware needs, and priorities.
Design self-service workflows that empower researchers to easily launch, observe, debug, and reproduce distributed workloads.
Boost infrastructure efficiency by enhancing GPU utilization, reducing scheduling latency, speeding up workload startup times, and maximizing throughput.
Engineer robust reliability measures, including observability tools, failure recovery processes, and capacity planning for critical machine learning tasks.
Participate in on-call rotations to troubleshoot and resolve issues spanning applications, schedulers, networking, storage, and GPU hardware.

Requirements

Bring four or more years of professional experience in machine learning infrastructure, distributed systems, Kubernetes platform engineering, or closely related fields.
Demonstrate strong proficiency in Python or Go when building and maintaining production-grade distributed systems.
Possess deep knowledge of Kubernetes, including controllers, operators, CRDs, advanced scheduling, networking, storage, and resource management.
Understand critical ecosystem technologies such as Kueue, Karpenter, Volcano, and Kyverno for workload scheduling, provisioning, and policy enforcement.
Comprehend distributed machine learning workloads, covering training, fine-tuning, evaluation, checkpointing, and batch inference.
Familiarity with GPU infrastructure elements like PyTorch, CUDA, NCCL, and high-performance networking.
Know core scheduling concepts including quotas, priorities, preemption, gang scheduling, topology awareness, and workload admission.
Demonstrate the capability to diagnose complex performance and reliability bottlenecks across software, orchestration, networking, storage, and hardware.
Care deeply about developer experience and enjoy translating intricate infrastructure into simple, reliable user interfaces.
Thrive effectively in ambiguous, fast-paced environments influenced by cutting-edge frontier AI research.

Skills and Tools

Python
PyTorch
Kubernetes
Go
CUDA
NCCL
Kueue
Karpenter
Volcano
Kyverno

About the Company

Mistral AI delivers full-stack artificial intelligence solutions ranging from frontier models to developer tools, client applications, and compute power. The organization partners with global enterprises tackling difficult challenges across high-stakes industries such as finance, manufacturing, defense, healthcare, and the public sector to co-create tailored AI systems. The dynamic, low-ego, and globally distributed team consists of creative individuals who are passionate about transforming society through responsible innovation.

How to Apply

Interested candidates should visit the official Mistral AI careers page to submit their application and review any additional location-specific details.

Skills and tech stack

PythonPyTorchKubernetes

Questions candidates ask

What is the primary focus of the Research Engineer, ML Platform role?

The role focuses on building and operating the machine learning platform that powers large-scale training, evaluation, and batch inference at Mistral AI.

How many years of experience are required for this position?

Candidates must have four or more years of experience in ML infrastructure, distributed systems, Kubernetes platform engineering, or a related field.

What programming languages are required?

Applicants must be proficient in Python or Go and comfortable working with production-grade distributed systems.

Where is the position located?

The role is based in Palo Alto with remote work options available as part of Mistral AI's distributed team setup.

How can I apply for this job?

Candidates can apply directly through the official Mistral AI company careers page.

Ready to apply?

The application is on the official Mistral AI careers page.

Apply on company site