Software Engineer, Collective Communication

OpenAI
San Francisco
Workplace: HybridFull timeUSD 380,000 - 555,000 annuallyFunction: Software EngineeringSkills: ["Collaboration","Teamwork","Problem-solving"]

Software Engineer specializing in low-level, performance-critical networking for large-scale AI training workloads. Design and implement custom networking collectives in C++/CUDA, collaborating with ML researchers, and contribute to simulations and network design for our supercomputers.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
OpenAI
OpenAI
2 years ago

Software Engineer, Collective Communication

āœ“ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 15 minutes agoStatus: Live

Job Summary

Software Engineer specializing in low-level, performance-critical networking for large-scale AI training workloads. Design and implement custom networking collectives in C++/CUDA, collaborating with ML researchers, and contribute to simulations and network design for our supercomputers.
Location: San Francisco
Workplace: Hybrid
Employment Type: Full time
Job Function: Software Engineering

Key Responsibilities

  • ā€¢Design and implement custom networking collectives that are tightly integrated into the training stack.
  • ā€¢Collaborate with ML researchers to design and implement efficient collective operations in C++ and CUDA.
  • ā€¢Ensure the largest training jobs leverage different network transports used on our supercomputers.
  • ā€¢Work on simulations to inform future supercomputer network designs.

Pay and Benefits

Salary: USD 380,000 - 555,000 annually
Equity and Bonus:Equity
Perks:RelocationEquity

Key Requirements

  • ā€¢Background in low-level, performance-critical software (C++ and CUDA) for distributed systems.
  • ā€¢Experience with distributed algorithms using RDMA is a plus.
  • ā€¢Ability to design and implement custom networking collectives tightly integrated into a training stack.
  • ā€¢Familiarity with network simulation techniques and CPU/GPU performance-sensitive code.
  • ā€¢Experience collaborating with researchers and engineers on high-performance computing workloads.
Experience:High performance computingAI researchDistributed systems
Skills:CollaborationTeamworkProblem-solving
Tech Stack:C++CUDARDMACPUGPUNetwork simulation

Company Brief

OpenAI
Develops and deploys advanced generative AI models (including ChatGPT and DALLĀ·E) and AI infrastructure, providing APIs and consumer products to accelerate safe AGI for broad benefit.
Industry: AI & Machine Learning
Company Size: Enterprise (1,001+ employees)
Revenue: USD 1B+
Growth: Scaleup
Valuation: Hectocorn (USD 100B+)
Funding: Series E+
Headquarters: San Francisco, United States
Founded: 2015
Glassdoor
Glassdoor: 4.4
WebsiteLinkedInGlassdoor