Production AI Engineer - Build Real-World Agent Tools

Garuda Ventures; San Francisco (CA); On-site; USD 190,000 - 240,000; Full time

Job summary

Garuda Ventures is building the next generation AI actions platform that lets agents securely take actions inside enterprise systems. You’ll own the orchestration of agent tooling, sub-agents, and tool availability, working with experts from Okta, Redis, Microsoft, and more.

The role focuses on end-to-end delivery of scalable APIs, tooling, and evaluation systems for AI agents, with early-stage startup dynamics and opportunities to shape product direction.

Qualifications

• 4+ years of software engineering experience shipping production code and owning projects end-to-end.
• Proficiency in Python and/or TypeScript.
• Experience building and consuming APIs at scale: REST, auth flows, pagination, rate limits, and real-world integrations.
• Experience with LLM applications, prompting, tool use, or agent design.
• A testing mindset with tests and evals that catch real failures beyond green checkmarks.
• Comfort with ambiguity and ability to operate in a early-stage team.

Responsibilities

• Build agent harnesses that write and maintain tools, plus agents-as-tools (sub-agents).
• Deliver jobs end-to-end, influencing Arcade Engine, MCP, or new services.
• Create toolkits and abilities from vendor APIs or internal needs to achieve real outcomes.
• Identify and fix rough edges to enable agents across customers.
• Maintain and improve the existing catalog; adapt to upstream API changes.
• Develop evaluation and testing for abilities to measure tool effectiveness across agents and surfaces.
• Engage with and advance the MCP ecosystem and tool standards.

Skills

4+ years experience

Python/TypeScript

APIs at scale

LLM applications

Testing mindset

Ambiguity tolerance

Job description

Garuda Ventures is building the next generation AI actions platform that lets agents securely take actions inside enterprise systems. You’ll own the orchestration of agent tooling, sub-agents, and tool availability, working with experts from Okta, Redis, Microsoft, and more.

The role focuses on end-to-end delivery of scalable APIs, tooling, and evaluation systems for AI agents, with early-stage startup dynamics and opportunities to shape product direction.