SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Anysphere, maker of Cursor (an AI-native code editor), is seeking a Software Engineer for the RL Data team to design and build tools that researchers and external contributors use to create, review, submit, and monitor environments and tasks for reinforcement-learning training runs.
This is a full-stack product-engineering role embedded in a research team. You will own the review and acceptance experience end-to-end: from rollout and transcript inspection, task-quality signal grading and reward-hacking analysis, to workflows that move submissions into training. You'll also build authoring interfaces enabling researchers, vendors, and domain experts to create and improve environments and tasks quickly and confidently.
Key responsibilities include:
- Creating fast, trustworthy workflows for vendors and the research team to collaborate effectively on task creation, iteration, and submission
- Building review tools for inspecting and comparing rollouts, transcripts, grader outputs, and other task-quality signals
- Developing environment-health, failure-search, versioning, and catalog experiences that make training data easy to understand, manage, and extend
- Establishing a shared component kit and self-serve interfaces for creating and improving tasks with inline quality checks
The company emphasizes a flat organization with a small, talent-dense team that values truth-seeking, passion, creativity, spirited debate, and shipping code.
REQUIREMENTS:
- Shipped full-stack products and owned systems from user interface through storage or services, using TypeScript and React alongside Node, Python, or Go
- Built dense, data-facing tools such as transcript viewers, diffing systems, review queues, observability products, or operational dashboards with strong opinions on structured data rendering
- Built or maintained a design system or component library and can establish durable product and engineering conventions for fast-moving teams
- Designed review, QA, moderation, fraud, or acceptance workflows where users had incentive to circumvent checks, with experience keeping such systems honest
- Care about data quality and willingness to inspect raw data; experience with evaluations, graders, reinforcement learning, or data-quality systems is helpful but not required
- Ability to move quickly under ambiguity, collaborate closely with researchers and domain experts, and take open-ended problems from rough need to reliable product
About Anysphere
AI / Data / Infrastructure — maker of Cursor, an AI-native code editor for software developers.
