Anthropic
Software Engineer, Tokens and Prompt Structures
About this role
Join Anthropic's Encodings Infra team to design and maintain the libraries that convert text and multimodal data into forms Claude can understand. You'll own the APIs and abstractions that enable researchers and engineers across the organization to work with encodings and prompt structures, while ensuring new research ideas move seamlessly from experimentation to production.
What you'll do
- Maintain and improve encoding libraries used across Anthropic by researchers and engineers
- Design experiments to determine optimal ways to feed structured data into Claude
- Create abstractions that simplify encoding complexity for most users while supporting advanced use cases
- Adapt encoding libraries to support new research directions and enable production deployment
- Optimize encoding performance across dependent systems throughout the codebase
What they're looking for
- Library and SDK design with 5+ years experience
- ML terminology and LLM architecture understanding
- Large codebase refactoring
- Python and/or Rust
- Performance optimization
- Tokenizers and text/data encoding systems
- Cross-team communication and collaboration
- Reinforcement learning or model training infrastructure
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Anthropic
Anthropic builds Claude, an AI assistant, and is hiring for engineering roles across infrastructure, data systems, and security that support both AI research operations and the company's internal technology needs. The company seeks infrastructure engineers, systems integrators, data scientists, and security specialists to build production-scale systems for training data pipelines, financial operations, developer productivity measurement, research infrastructure, and server firmware security.
- Website
- anthropic.com
Likely interview questions
- Describe your experience maintaining a widely-used library or SDK. How did you balance breaking changes with backwards compatibility?
- Tell us about a complex refactor you led in a large codebase. How did you manage coordination across multiple teams?