Responsibilities
- Design and build distributed failure detection, tracing, and profiling systems for large-scale AI training jobs.
- Develop tooling to identify slow, faulty, or misbehaving nodes and provide actionable system visibility.
- Improve observability, reliability, and performance across the training platform.
- Debug and resolve issues in complex, high-throughput distributed systems.
- Collaborate with systems, infrastructure, and research teams to evolve platform capabilities.
- Extend failure detection and tracing systems for new training paradigms and workloads.
Requirements
- Experience writing low-level software where system details matter.
- Understanding of hardware, operating systems, networking, concurrency, and distributed systems.
- Background in high-performance computing or low-level systems engineering.
- Interest in performance, stability, observability, and automating operational workflows.
About OpenAI
OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. AI is an extremely powerful tool that must be created with safety and human needs at its core. OpenAI is dedicated to putting that alignment of interests first — ahead of profit. To achieve our mission, we must encompass and value the many different perspectives, voices, and experiences that form the full spectrum of humanity. Our investment in diversity, equity, and inclusion is ongoing, executed through a wide range of initiatives, and championed and supported by leadership. At OpenAI, we believe artificial intelligence has the potential to help people solve immense global challenges, and we want the upside of AI to be widely shared. Join us in shaping the future of technology.