About the position
• Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.
• As a Prompt Engineer on the Claude Code team, you'll own Claude's behaviors specifically within Claude Code — ensuring users get a consistent, safe, and high-quality experience as we ship new models and evolve the product. This is a highly specialized role sitting at the intersection of model behavior and product quality.
• You'll be the expert on how Claude behaves inside Claude Code, owning and maintaining the system prompts that ship with each new model snapshot. When a new model drops, you're the person making sure Claude Code feels right within days — not weeks. You'll work closely with Model Quality and Research to understand emergent behaviors and behavioral regressions, and with product and safeguards teams to respond quickly when something goes wrong.
• This role requires someone who can move fast on behavioral tuning while maintaining rigor, and who cares deeply about the end-to-end developer experience Claude Code delivers. You'll need strong prompting skills, excellent judgment about model behaviors, and the collaborative skills to work across product, safeguards, and research teams.
• Salary: $320,000-405,000 (SWE-G 5-6)
• Responsibilities
• Own Claude Code's system prompts for each new model snapshot, ensuring behaviors feel consistent and well-tuned
• Review production prompt changes and serve as a resource for particularly challenging prompting problems involving alignment and reputational risks
• Lead incident response for behavioral and policy concerns, coordinating with product and safeguards teams
• Scale prompting and evaluation best practices across claude code and product teams.
• Deliver product evaluations focused on model behaviors
• Define and streamline processes for rolling out prompt changes, including launch criteria and review practices
• Create model-specific prompt guides that document quirks and optimal prompting strategies for each release
• Collaborate with product teams to translate feature requirements into effective prompts
• Requirements
• Are a power user of agentic coding tools and have strong intuition about model capabilities and limitations
• Thrive in high-intensity environments with fast iteration cycles
• Take full ownership of problems and drive them to completion independently
• Are skilled at creating and maintaining behavioral evaluations
• Have strong technical understanding, including comprehension of agent scaffold architectures and model training processes
• Are an experienced coder comfortable working in Python and Typescript
• Have independently driven changes through production systems with strong execution and responsiveness
• Have experience translating user feedback and product needs into coherent prompts and behavioral specifications
• Excel at working across organizational boundaries, collaborating effectively with teams that have differing goals and perspectives
• Have experience translating user feedback and behavioral observations into coherent prompt changes and specifications
• Care deeply about AI safety and making Claude a healthy alternative in the AI landscape
• We require at least a Bachelor's degree in a related field or equivalent experience.
• Benefits
• competitive compensation and benefits
• optional equity donation matching
• generous vacation and parental leave
• flexible working hours
• a lovely office space in which to collaborate with colleagues
Apply tot his job
Apply To this Job