Software Engineer, Tokens and Prompt Structures
About this role
Anthropic needs to bridge the gap between complex machine learning research and practical production systems by maintaining the core libraries that encode text and multimodal data for Claude. The Encodings Infra team solves this by defining prompt shapes, tool calling representations, and data structures so that the rest of the organization can build features without handling raw encoding details.
Software engineers on this team own the design, performance, and maintenance of these critical libraries while collaborating closely with researchers to move experimental ideas into production. Successful candidates bring strong software engineering foundations, a collaborative approach to working with research teams, and the capability to execute complex codebase refactors.
What you will do
- Maintain and improve the encoding libraries used by engineers and researchers across Anthropic.
- Run experiments to determine the optimal way to feed structured data into Claude without confusing it.
- Design data structures and abstractions that shield most of the organization from the details of how encoded data works while enabling power users.
- Adapt the encoding libraries to support new research directions as they emerge and ensure successful production deployment.
- Optimize encoding performance across the systems that depend on these libraries.
What you need
- Have 5+ years of software engineering experience, with meaningful time spent maintaining libraries, SDKs, or developer-facing APIs.
- Possess familiarity with machine learning terminology and large language model architecture to work effectively alongside researchers.
- Demonstrate experience carrying out complex refactors in large codebases.
- Exhibit strong communication skills and enjoy working closely with researchers and engineers.
- Maintain a results-oriented approach with a bias towards flexibility, impact, and stepping in outside standard job descriptions.
Nice to have
- Tokenizers or other text/data encoding systems
- Maintaining a widely-used library over a long period of time
- Performance optimization
- Python and/or Rust
- Reinforcement learning or model training infrastructure
Benefits
- Competitive compensation and benefits
- Optional equity donation matching
- Generous vacation and parental leave
- Flexible working hours
Skills and tech stack
Questions candidates ask
What is the remote or hybrid work policy?
Anthropic operates under a location-based hybrid policy. All staff are expected to be in one of the offices at least 25% of the time, though some roles may require more time in the office.
What is the annual compensation range for this role?
The annual base salary range for this position is $320,000 to $405,000 USD.
Does the company provide visa sponsorship?
Yes, Anthropic does sponsor visas and retains an immigration lawyer to assist with the process. If an offer is extended, the team makes every reasonable effort to secure a visa.
What kind of background do I need in machine learning?
Candidates do not need to be machine learning experts. However, familiarity with ML terminology and LLM architecture is necessary to work effectively alongside researchers and understand experiment results.