About the Team

The Applied team safely brings OpenAI's technology to the world. We released the GPT-3 API and Codex, which powers GitHub's Copilot. There's a lot more on the immediate horizon.

Our customers build fast-growing businesses around our APIs, which power product features that were never before possible. We simultaneously ensure that our powerful tools are used responsibly. Safe deployment is more important to us than unfettered growth.

The Engineering team wraps a massive fleet of GPUs in scalable, robust, infrastructure powered by Kubernetes, Go, Python, Terraform, Redis, Kafka, Postgres, and Snowflake. Our APIs are powered by Python Flask and OpenAPI with a React frontend.

About the Role

We are looking for an engineer who wants to take the world's largest and most capable AI models and optimize them for use in a high-volume, low-latency, and high-availability production environment.

In this role, you will:

• Optimize our code and fleet of Azure VMs to utilize every FLOP and every GB of GPU RAM of our hardware.
• Build tooling to give us visibility into our bottlenecks and sources of instability and then architect and implement solutions to address the highest priority issues.
• Introduce new techniques, tools, or architectures that improve the performance, latency, throughput, and efficiency of our models.
• Work alongside machine learning researchers, engineers, and product managers to bring our latest technologies into production.

You might thrive in this role if you:

• Have at least 3 years of professional software engineering experience.
• Have an understanding of modern ML architectures and an intuition for where optimizations can be made.
• Have or can quickly gain familiarity with PyTorch, NVidia GPUs and the software stacks that optimize them (e.g. NCCL, CUDA), as well as HPC technologies such as InfiniBand, MPI, etc.
• Have experience architecting, observing, and debugging production distributed systems.
• Have a humble attitude, an eagerness to help your colleagues, and a desire to do whatever it takes to make the team succeed.
• Have needed to rebuild or substantially refactor production systems several times over due to rapidly increasing scale.
• Are self-directed and enjoy figuring out the most important problem to work on.
• Own problems end-to-end, and are willing to pick up whatever knowledge you're missing to get the job done.
• Build tools to accelerate your own workflows, but only when off-the-shelf solutions would not do.