Software Engineer, Habitat (Online Data)
OpenAIAbout the team
Online Data builds and operates Habitat, the single product surface of Online Data and the system of record for OpenAI’s online user data. As OpenAI’s scale and product requirements evolve, Habitat is becoming a full-stack, one-size-fits-most database platform with end-to-end ownership of:
Provisioning and developer experience
APIs and guardrails
Scaling, performance, and reliability
Data movement, caching, routing, and placement
Privacy enforcement and access control
Change Data Capture (CDC) as a first-class primitive
The foundation for future storage backends
You’ll work on the core online database platform behind OpenAI’s products, building and operating Habitat services that handle high-QPS, latency-sensitive workloads across regions. You’ll partner closely with internal platform and product teams to ship safe, reliable systems, then push them to be faster and more cost-efficient through better caching, routing, observability, and operational tooling. This is a critical role for engineers who like owning hard distributed-systems problems end to end and sweating the details from p99 latency to production operations at massive scale.
In this role, you will
Design and build core abstractions spanning storage, caching, routing, CDC, and privacy enforcement
Own a major surface area end to end, from product and API design to operational excellence
Improve latency, correctness, and cost efficiency for real production workloads at massive scale
Build strong instrumentation, debugging workflows, and developer-first tooling
Collaborate closely with internal product and infrastructure teams to understand requirements and ship pragmatic solutions
Participate in an on-call rotation and raise the bar on reliability while aggressively improving performance and usability
You might thrive in this role if you have
A strong track record building and operating high-scale backend or data-intensive distributed systems in production
Excellent systems judgment and the ability to make tradeoffs across latency, cost, correctness, and reliability