AI Operations Engineer

The AI Operations Team

The AI Operations team owns the agentic layer the rest of the company runs on. We build the workflows, agents, and scheduled tasks that automate business processes end to end, and we build and maintain the proprietary framework those components ship in. Every business unit at OpenZeppelin is a customer of this team.

What youll be doing

The role has two primary work streams.

AI workflow design, implementation, and maintenance. You understand the fundamental AI building blocks that we use in the company, and you work with interdisciplinary business units to help them identify flows in their processes that can be optimized and automated with AI.

Internal AI automation framework product feedback and contribution. Based on the experience that you obtain as a consumer of the internal AI automation framework, and on your study of the ecosystem, you provide product feedback (feature requests, bug reports, etc.), contribute to the source code of the framework, and take part in peer review and development.

• Work together with different business units in the company to understand where and how AI can be applied in their daily workflows to improve their overall productivity.

• Design, implement, test, and maintain AI automation workflows and scheduled tasks that encapsulate and automate business processes that span the company.

• Monitor and measure automation executions, and iterate on the components that underperform.

• Provide product feedback for fixes and new features in the internal AI automation framework, and peer review and actively maintain its source code as a developer on the AI Operations team.

• Run training sessions and write internal guides on applied AI topics such as skill development, agent design, and prompt/context engineering.

  You have

• Hands-on experience building applied AI systems, and fluency with both the concepts involved (prompt engineering, context engineering, prompt caching, tool use, progressive disclosure) and the fundamental building blocks: skills, remote and local MCP servers, agents, and agent harnesses.

• Practical experience with LLM provider APIs and gateways, such as the the OpenAI Chat Completions and Responses APIs, the Anthropic Messages API, or OpenRouter.

• Ability to read, write, and structure Markdown-based systems (skills, agent definitions, workflow specs, and knowledge bases), alongside TypeScript development experience and solid Git and GitHub fluency: branches, pull requests, reviews, and the discipline to keep shared repositories clean.

• A security-first mindset: you think about what an autonomous agent should

  not

  be allowed to do before you ship it.

• Strong ability to communicate deeply technical topics, both written and spoken, and patience working with non-technical colleagues. You will be their bridge into the agentic workflow.

  Nice to have

• Familiarity with AI software development techniques like spec-driven development or RPI (Research, Plan, Implement), with the principal AI development harnesses like Claude Code, Codex, or OpenCode, and with SDD tooling like Superpowers or Spec Kit.

• Experience running evaluations of LLM systems (evals, benchmarks, regression testing of prompts).

• Ability to work autonomously and as a collaborator on the team.

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.