Responsibilities

Build and operate autonomous agents for inbound, outbound, pipeline management, and customer engagement workflows.
Design approval gates, handoffs, and escalation paths that maintain appropriate human oversight.
Develop and run agent evaluation frameworks in development and production.
Instrument model and tool calls and build observability and measurement connecting agent actions to pipeline and revenue.
Ship MCP servers, agent skills, and web applications integrating with CRM, communication, and data warehouse systems.
Set technical direction for agent development, evaluation, and operations across the team.
Architect shared codebases and establish contribution, convention, testing, and review practices.
Work directly with sellers to ground agent designs in real workflows and iterate from observed results.
Identify repeatable patterns and share insights with Product and Engineering teams.

Requirements

Strong programming skills in Python or TypeScript and experience building and operating production applications.
Production experience with LLMs, context engineering, agent development, MCP development, tool use, and evaluation frameworks.
Experience using evaluations and transcript analysis to identify and resolve problems in LLM systems.
Working fluency with data, including SQL.
Ability to navigate ambiguity and ship without a detailed specification.
Passion for safe and beneficial AI and consideration for the people using the systems built.
Preferred: 8+ years in software engineering, ML engineering, or forward deployed engineering roles.
Preferred: Experience with Claude Code and the Claude Agent SDK.
Preferred: Experience with CRM, sales engagement, enrichment, conversation intelligence, or revenue teams.
Preferred: Experience growing codebases with many contributors, including inner-source or open-source environments.
Preferred: Applied ML and experimentation experience involving A/B testing, propensity models, recommendations, or causal analysis.
Preferred: Exceptional communication skills for explaining technical concepts to non-technical partners.

Hybrid policy requiring staff to work from an office at least 25% of the time, with some roles requiring more office time.
Visa sponsorship with immigration-lawyer support when applicable.
Competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and office collaboration space.
Anthropic is headquartered in San Francisco and is a public benefit corporation.