Software Engineer, Agent Experience
$160k - $210k • Remote • North America
Posted 10d ago
Job Location
North America
Tech Stack
Remote Work Policy
Fully remote
Categories
AI Agent Engineer
About the job
AssemblyAI is seeking a Software Engineer to own the agent experience (AX) for developers building with their Voice AI models. This role focuses on ensuring AI coding agents can seamlessly discover, integrate, and debug AssemblyAI products. You will be responsible for developing agent skills, prompts, and integrations that enable agents to create production-grade AssemblyAI integrations on the first try. This is a unique opportunity to shape a new and rapidly growing area within a successful AI company, working closely with senior engineers and gaining significant ownership from day one.
Responsibilities
- Drive the agent experience, treating coding agents as a primary user.
- Build and maintain agent skills for core products like Speech-to-Text, Real-time, Dictation, Voice Agents, TTS, and LLM Gateway.
- Develop and extend the AssemblyAI MCP server and agent integrations, including authentication.
- Make documentation and code samples agent-ready, ensuring agents select the correct APIs and patterns.
- Tune prompts and instructions for in-dashboard agents and onboarding flows.
- Build evaluations and benchmarks to measure agent success rates and integration times.
- Scale the agent integration playbook across all products.
- Utilize AI tools in daily work and help other engineers leverage agents for faster development.
Requirements
- 2+ years of professional software engineering experience with strong full-stack fundamentals in Javascript and/or Python.
- Daily, hands-on experience with coding agents like Claude Code, Cursor, or Codex.
- Experience writing prompts, skills, tool definitions, or agent instructions, with an interest in measuring their effectiveness.
- Strong product sense for developer-facing products, focusing on successful integration.
- Clear technical writing skills for both agents and humans.
- Comfort with ambiguity and a bias towards shipping new features.