Engineering · Full-time · San Francisco, New York, Remote
Our mission is to automate coding. The first step in our journey is to build the best tool for professional programmers, using a combination of inventive research, design, and engineering. Our organization is very flat, and our team is small and talent dense. We particularly like people who are truth-seeking, passionate, and creative. We enjoy spirited debate, crazy ideas, and shipping code.
### About the role
We’re hiring a Software Engineer, Reliability to help Cursor scale with a high reliability bar.
You will work across the stack (client, backend services, model routing/integrations, and infra) to find the reliability bottlenecks that most impact users, ship durable fixes, and build the tooling and guardrails that keep us fast and stable.
This is a strongly engineering-focused role. It is not a program management role, and it is not an ops-only SRE role.
### What you’ll do
- Own reliability work end-to-end, from user-facing symptoms (crashes, latency, streaming failures) to root causes in services, infrastructure, or vendor dependencies. - Design and implement resilience patterns for upstream dependency failures (for example model providers): fallbacks, routing strategies, and degraded-mode designs. - Build and maintain reliability guardrails that make teams faster and safer: deployment safety, rollbacks, operational playbooks, automated checks, and standards for production readiness. - Improve observability (metrics, logs, traces, and client telemetry) so engineers can quickly answer “Is it up?” and “What changed?” - Reduce operational toil through automation and better tooling. - Partner with product and infrastructure engineering teams as a drop-in reliability multiplier: embed on the highest-impact problems and drive them to a durable technical outcome. - Participate in an on-call rotation and help improve incident response practices over time (severity definitions, runbooks, retrospectives, and clear ownership of follow-up fixes).
- You will own a small set of high-leverage reliability “themes” at a time (for example client crash rate, streaming reliability, deploy safety). You drive these end-to-end until the reliability bar measurably moves. - You will not be “responsible for everyone’s metrics” by default. You will build the system and partner with teams; service owners ultimately own their service SLOs and fixes. - You will not be the owner of all CI/CD. You will raise the production-readiness bar with guardrails and tooling, while infrastructure and product teams own their pipelines and day-to-day workflows. - On-call is part of the job, but the goal is to eliminate recurring incidents and toil, not to be a permanent triage function.
### You may be a fit if
- Have a track record of improving reliability by empowering other engineers with excellent tooling, guardrails, and simple operational systems. - Own problems end-to-end, learn quickly, and enjoy working across layers (client symptoms, service behavior, infra primitives, and third-party dependencies). - Prefer pragmatic, high-leverage fixes over perfection, and can raise standards without becoming “the voice of no.” - Are comfortable leading through influence: aligning teams on the “why,” landing changes in multiple codebases, and driving clarity on ownership.
- Strong experience owning reliability for production systems, including both incident response and long-term engineering fixes. - Strong software engineering instincts. You write code to automate, eliminate recurring operational work, and prevent regressions. - Expert-level experience in at least one of: Go, Node/TypeScript, or Python. - Deep practical knowledge of cloud infrastructure (AWS) and modern deployment/orchestration patterns (Kubernetes and/or ECS). - Experience with observability systems and practices (metrics, logs, traces, and alerting). - Clear communication and cross-team leadership.
### Bonus points
- Experience with multi-region architecture and global distribution strategies. - Experience with networking and long-lived connection workloads (for example HTTP/2 streaming). - Experience building reliability programs in high-growth orgs with incredibly high velocity.
#LI-DNI