AI Engineer (Agents Lead)
Tambo • Remote (United States)
Posted: January 31, 2026
Job Description
AI Engineer (Agent Lead)
Location: Remote (Pacific Time preferred)
Type: Full-time
About Tambo
Tambo is an open-source toolkit for building agents that render your React components. We're five people, pre-seed, building infrastructure for AI-native apps. Everyone operates like a mini-founder for their area.
The Role
You'll own the agent systems that make Tambo work. That means building the runtime, improving how agents reason and act, and figuring out what's possible when AI meets visual interfaces.
We're not looking for someone who's "interested in agents." We want someone who's built them.
What You'll Do
Build agent architectures that are fast, reliable, and actually pleasant to work with
Research and ship techniques for agent reasoning, planning, and tool use
Create evaluation pipelines so we can measure whether agents are getting better or worse
Talk to users. Understand what's broken. Fix it.
What We're Looking For
5+ years shipping software
You've actually built agents or LLM-powered apps (not just played with them)
You understand what LLMs can and can't do
You're comfortable with ambiguity. We're five people. There's no playbook.
Bonus Points
Experience with TypeScript
Open-source or developer tools background
You've written or published stuff about agents
You've built or worked on coding agents
Comp may shift based on location and experience.
Additional Content
AI Engineer (Agent Lead)
Location: Remote (Pacific Time preferred)
Type: Full-time
About Tambo
Tambo is an open-source toolkit for building agents that render your React components. We're five people, pre-seed, building infrastructure for AI-native apps. Everyone operates like a mini-founder for their area.
The Role
You'll own the agent systems that make Tambo work. That means building the runtime, improving how agents reason and act, and figuring out what's possible when AI meets visual interfaces.
We're not looking for someone who's "interested in agents." We want someone who's built them.
What You'll Do
Build agent architectures that are fast, reliable, and actually pleasant to work with
Research and ship techniques for agent reasoning, planning, and tool use
Create evaluation pipelines so we can measure whether agents are getting better or worse
Talk to users. Understand what's broken. Fix it.
What We're Looking For
5+ years shipping software
You've actually built agents or LLM-powered apps (not just played with them)
You understand what LLMs can and can't do
You're comfortable with ambiguity. We're five people. There's no playbook.
Bonus Points
Experience with TypeScript
Open-source or developer tools background
You've written or published stuff about agents
You've built or worked on coding agents
Comp may shift based on location and experience.