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 building systems to train AI models in realistic, diverse environments. The company's mission is to automate coding, and one north star is enabling teams to hire AI agents as remote colleagues.
As a Software Engineer on the RL Environments team, you will build the infrastructure that transforms real-world data into high-quality training environments and tasks for reinforcement learning. You'll own platforms and quality layers that help produce trustworthy data at scale, working at the intersection of research, data engineering, and ML infrastructure.
Key responsibilities:
- Build platforms that create complex, realistic, and diverse RL environments at scale, supporting end-to-end tasks across knowledge-work domains
- Design end-to-end data factories that convert massive raw data into useful, realistic training environments
- Define and apply consistent quality standards across vendor, tutor, acquired, and synthetic data sources
- Create self-serve APIs and tooling that accelerate task and environment development for internal teams and external contributors
- Partner across research, platform, and external teams to translate model-capability goals into effective training tasks and experiments
The organization is flat, small, and talent-dense. The team values truth-seeking, passion, creativity, spirited debate, and shipping code.
REQUIREMENTS:
- Strong software engineering fundamentals and experience with data platforms, developer tools, distributed systems, or ML infrastructure
- Ability to turn ambiguous quality standards into concrete, automated checks
- Comfort building repeatable pipelines from messy, heterogeneous data
- Understanding of model behavior and ability to translate capability goals into tasks and experiments
- Strong cross-disciplinary collaboration skills and ability to own open-ended problems end-to-end
About Anysphere
AI / Data / Infrastructure — maker of Cursor, an AI-native code editor for software developers.
