Software Engineer, Agent Infrastructure
San Francisco • FullTime
Posted 1y ago
Remote Work Policy
On-site
Employment Type
FullTime
Categories
AI Agent Engineer
About the job
The Agent Infrastructure team at OpenAI is building systems to enable the training and deployment of highly useful AI agents. This role involves working closely with research and product teams to scale systems for training agentic models and to build the platform for launching new agents to millions of users. You will contribute to a novel container orchestration platform, develop and maintain APIs for agentic infrastructure, and use Terraform for infrastructure management. The ideal candidate has deep experience in large-scale machine learning infrastructure, a knack for performance optimization, and expertise in virtualization and containerization technologies.
Responsibilities
- Contribute to a novel, in-house container orchestration platform to scale beyond Kubernetes.
- Develop and maintain FastAPI and gRPC APIs for agentic infrastructure in training and production.
- Utilize Terraform to establish and evolve complex infrastructure for research and production.
- Collaborate with research teams to set up and optimize systems for AI training runs and experimental applications.
Requirements
- Deep experience working on large-scale machine learning infrastructure.
- Ability to reason about training at scale, identify bottlenecks, and engineer solutions.
- Experience building new systems from scratch and scaling them significantly.
- Keen eye for performance and optimization in complex, globally-distributed systems.
- Proficiency with cloud platforms and infrastructure-as-code technologies like Terraform.
- Experience with virtualization and containerization technologies (e.g., Kata, Firecracker, gVisor, Sysbox).
- Passion for optimizing runtime performance.
Benefits
- Hybrid work model (3 days in office per week).
- Relocation assistance for new employees.