B

Baseten

Runs AI models for production software

Software Engineer - Training Infrastructure

Platform EngineeringConfirmed live in the last 24 hours
Full-timeHybrid

About the Job

Requirements

  • A bachelor's degree or higher in Computer Science or a related field.
  • At least 5 years of experience.
  • Proficiency in Go.
  • Deep expertise with Kubernetes in production environments.
  • Advanced understanding of distributed systems concepts and performance tuning.
  • Proven experience designing observability systems.
  • Experience with machine learning/artificial intelligence workloads and MLOps platforms.

Responsibilities

  • Design and architect scalable infrastructure systems for the machine learning training platform, including scheduling, storage, and networking.
  • Partner closely with developers and research engineers to translate complex training requirements into technical solutions.
  • Design and architect a global training scheduler.
  • Design and architect reinforcement learning systems and continuous learning pipelines.
  • Drive long-term improvements to increase system reliability and development velocity.
  • Partner closely with SRE and Capacity teams to enable state-of-the-art training infrastructure.
  • Make critical architectural decisions that balance performance with system reliability.
  • Lead technical discussions and mentor junior engineers on infrastructure best practices.
  • Contribute to long-term technical strategy and the infrastructure roadmap.

Desired Qualifications

  • Experience with distributed storage systems.
  • Python experience.
  • Extensive experience with major cloud providers such as AWS and GCP, and neo-cloud providers such as Crusoe, DigitalOcean, and Nebius.
  • Experience with workload orchestration platforms such as Temporal or Airflow.
  • Familiarity or experience with the open-source training stack and frameworks, including NCCL, PyTorch, Megatron, NemoRL, VeRL, Axolotl, and HF Trainer, and distributed training techniques such as FSDP and DeepSpeed.
  • Experience developing AI products, tooling, or agents.

About the Company

Baseten

Runs AI models for production software

Baseten runs AI models for software companies that need fast, reliable responses inside their products, from text generation to speech and image processing. Its Inference Stack combines optimized model runtimes with infrastructure that deploys, serves, and monitors custom or open models in production. Teams package models with its open source Truss tool, while Model APIs let developers test ready-to-use models and Training runs jobs that adapt models to their own data. Customers pay for computing time or tokens processed, with negotiated plans for larger deployments and support needs. Baseten was founded in 2019 and is headquartered in San Francisco.

Company size
201-500
Company stage
Series F
Total funding
$2.1B
Headquarters
San Francisco, CA
Founded
2019

Recently Posted Jobs

Explore more opportunities at Baseten

See more jobs