Software Engineer, Tokens and Prompt Structures
Anthropic · San Francisco, CA | New York City, NY
About this role
**Software Engineer, Tokens and Prompt Structures** **About Anthropic** Anthropic’s mission is to create reliable, interpretable, and steerable AI systems—safe and beneficial for users and society. **About the Role** The Encodings Infra team maintains the libraries used across Anthropic to encode text and multimodal data into a form Claude can consume. The team also determines Claude’s prompt shape—how user turns are represented, how tools are called, and how tool results are received. As a Software Engineer on this team, you’ll own the design and maintenance of these libraries: keeping APIs intuitive, performance sharp, and abstractions solid so most of the org never has to think about encodings or prompt structures. You’ll collaborate closely with researchers and engineers to move new encoding ideas from experiment to production. **Responsibilities** - Maintain and improve the encoding libraries used across Anthropic - Run experiments to determine the optimal way to feed structured data into Claude - Design data structures and abstractions that shield most of the org from encoding details while enabling “power users” - Adapt the libraries to support emerging research directions and ship them to production - Optimize encoding performance across dependent systems **You may be a good fit if you:** - Have 5+ years of software engineering experience, including meaningful time maintaining libraries/SDKs/developer-facing APIs - Understand ML terminology and LLM architecture well enough to work alongside researchers and interpret experiment results - Have experience with complex refactors in large codebases - Communicate clearly and enjoy working closely with researchers and engineers - Are results-oriented, with a bias toward flexibility and impact - Can pick up slack even outside your job description - Care about the societal impacts of your work **Strong candidates may also have experience with:** - Tokenizers or other text/data encoding systems - Maintaining a widely-used library over a long period - Performance optimization - Python and/or Rust - Reinforcement learning or model training infrastructure **Representative projects** - Working with a research team to ship a new multimodal data type (audio, video, etc.) to production - Redesigning a core abstraction so encoding changes don’t break downstream teams **Compensation (Annual Salary)** - **$320,000 — $405,000 USD** **Logistics** - **Minimum education:** Bachelor’s degree or equivalent combination of education/training/experience - **Required field of study:** Relevant to the role (via coursework, training, or professional experience) - **Minimum years of experience:** Correlates with internal job level requirements - **Location-based hybrid policy:** Expected to be in an office at least **25%** of the time (some roles may require more) - **Visa sponsorship:** Anthropic sponsors visas, though not for every role/candidate; reasonable efforts are made if offered **Apply even if you don’t meet every qualification** Not all strong candidates will match every listed requirement. If you’re interested in the work, you’re encouraged to apply. **Safety note** Anthropic recruiters only contact you from **@anthropic.com** email addresses (and vetted recruiting agencies may be used). Legitimate recruiters will never ask for money/fees/banking information before your first day. **How we’re different** Anthropic emphasizes “big science,” collaboration, and high-impact research with fre
Listing freshness
CronJobs last confirmed this listing 58m ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.