AI Engineer - Tools

About the role

Arcade is the MCP runtime that gives agents the power to securely take action inside enterprise systems. We connect agents to the systems they act in, then give each one a permission slip and a paper trail — proof of what it's allowed to do, and a record of what it did. Our solution is already shipping inside Fortune 100 companies.

As a Software Engineer on Tools, you'll report to the Engineering Manager for Tools and Growth. You'll build and maintain the toolkits agents depend on: take a vendor's API, design a clean set of tools on top of it, write the integration code, test it against real API behavior, and keep it healthy as those APIs change underneath us. You'll work close to the agent itself — how a model reads a tool description, how it decides which tool to call, and whether the result comes back in a shape the agent can use. You'll be doing all of this by building our own agentic tools to automate as much as this process as possible. The team is early and the catalog is growing fast, so you'll have real ownership from day one and a direct hand in the patterns every toolkit author after you inherits.

What You'll Do

• Build the agents we use internally to build, maintain, and scale our toolkits and skills.
• Build new toolkits — take a vendor's API and turn it into a high-quality set of tools that agents can call reliably.
• Maintain and improve the existing catalog — fix breakages, sharpen tool descriptions and response shapes, and keep quality high as upstream APIs change.
• Write tests and evals that prove a tool does what it claims — against real API behavior and across different models.
• Treat LLMs and agents as a first-class part of the job — tune tool descriptions, tool-use behavior, and how an agent actually calls what you built.
• Contribute to the shared patterns and tooling that make building the next toolkit faster.
• Engage with and push the MCP ecosystem, where the standards for agent tools are still taking shape.

Required Skills

• 4+ years of software engineering experience shipping production code.
• Strong Python and TypeScript.
• Experience building and consuming APIs — REST, auth flows, pagination, rate limits, and the messy parts of real-world integrations.
• A testing mindset — you write tests that catch real failures, not just green checkmarks.
• Comfort with ambiguity — early team, a narrow charter that will expand, decisions made with incomplete data.
• An insatiable desire to ship.

Bonus Points

• LLM application experience — prompting, retrieval, tool use, or agent design.
• Familiarity with the MCP (Model Context Protocol) ecosystem or similar agent-tool protocols — extra bonus if you've filed an issue against the spec.
• You've built evals or measurement systems for ML/AI behavior.
• Prior experience at an API platform, integrations-heavy product, or developer tools company.
• Open-source contributions.
• Experience at an early-stage startup, and you loved it.