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 roleAs a Software Engineer on ML Platform at Cursor, you'll build the infrastructure that turns real product usage into better models — and keeps research moving fast on large GPU fleets. ML Platform is organized into four teams. Depending on your background, you may join any of them:
Telemetry — Own the collection and serving path that turns real product use into a record research can trust; without slowing the product, and under a small, explicit policy. Client-side or high-volume ingestion experience is a plus.
ML Data Platform — Build the shared environments and pipeline substrate researchers extend, so new experiments don’t fork their own stack.
Observability — Make it easy for researchers to start, watch, and debug their own runs.
ML DevX and Systems — Shorten the path from idea to a trusted run on the research fleet.
We're looking for strong distributed-systems and infrastructure engineers who want to sit next to research and ship platform primitives that move the product.
We're in-person with cozy offices in North Beach, San Francisco, Palo Alto, and Manhattan, New York, complete with well-stocked libraries.
What you’ll doDesign, build, and operate core platform systems used daily by ML researchers and product engineers
Partner closely with research to turn recurring pain into durable infrastructure
Own reliability, performance, and developer experience for the systems in your lane
Ship iteratively in a flat, high-ownership environment. Measure impact, then raise the bar
You have a strong background in systems / infrastructure software engineering and enjoy building platforms other engineers depend on
You've owned production distributed systems at meaningful scale (ingestion, data pipelines, scheduling/orchestration, or similar)
You're comfortable across Linux, cloud and/or bare metal, and modern orchestration (Kubernetes, Ray, or equivalent)
You like working closely with ML researchers and product engineers
You thrive where ownership is high and the feedback loop is short
Telemetry: event ingestion, product analytics pipelines, OpenTelemetry / tracing, reliable data APIs
Product Data Platform: data frameworks, Spark / Flink / Ray, ML dataset and training-data infrastructure
Observability: experiment / run monitoring, debug and eval tooling, agent-friendly observability UX
ML DevX and Systems: GPU / cluster scheduling, job queues, node health, research compute developer experience
If there appears to be a fit, we'll reach out to schedule 2-3 short technicals. After, we'll schedule an onsite in our office, where you'll work on a small project, discuss ideas, and meet the team.
Skills Required
- Strong background in systems or infrastructure software engineering
- Experience building platforms that other engineers depend on
- Experience owning production distributed systems at meaningful scale, such as ingestion, data pipelines, scheduling, or orchestration systems
- Comfort working across Linux, cloud or bare-metal infrastructure, and modern orchestration systems such as Kubernetes or Ray
- Ability to work closely with ML researchers and product engineers
- Event ingestion, product analytics pipelines, OpenTelemetry, tracing, or reliable data APIs
- Data frameworks, Spark, Flink, Ray, or ML dataset and training-data infrastructure
- Experiment monitoring, run monitoring, debugging, evaluation tooling, or observability user experience
- GPU or cluster scheduling, job queues, node health, or research-compute developer experience
What We Do
We'd like to automate coding. To advance that mission, we're building Cursor. Our work includes training the world’s most widely used coding models, creating infrastructure that supports billions of requests per day, and building better ways for humans and AIs to work together.
