AI Operations Engineer

We're hiring our founding AI Operations Engineer. You'll be a founding member of this team, building the tools, integrations, and automations that make Pulley fit seamlessly into how our strategic customers work, plus the agents and automations that take manual work off our Permit Ops team. You'll sit on our Engineering team and work across Permit Ops, customers, and Product. You'll find the process that's slowing people down, build the thing that fixes it, and keep improving it until it works.

Responsibilities

Work with strategic customers, alongside the account leads, to understand how they run their rollouts and where Pulley needs to plug in; Build integrations, data syncs, and AI workflows that connect Pulley to customers' systems, safely; Sit with Permit Ops PMs, map how work actually gets done, and spot what's worth automating and what should stay human; Build fast, lightweight internal agents, tools, and automations that take manual research and admin work off their plate; Make sure teams actually adopt and trust what you ship; Build the evals and monitoring that tell us whether an AI workflow is getting it right; Turn one-off fixes into reusable tools and playbooks that work for every customer and every PM; Bring what you learn back to Product and Engineering so the platform gets better every week; Push back when a request won't actually solve the problem

Requirements

3+ years of experience in software engineering, ops engineering, applied AI, or a similar hands-on building role; CS or adjacent technical degree; A track record of shipping AI-powered tools, agents, or automations that are still running in production (a portfolio or examples are a big plus); Fluency with AI coding and agent tooling like Claude Code or Cursor as a daily practice; Solid engineering fundamentals: you can work with APIs and data, build reliable and secure systems, and debug things you didn't build; Experience working directly with customers or external stakeholders; Based in the San Francisco Bay Area and willing to work in person 4 days a week

Nice to have

Experience in construction tech, govtech, proptech, or another domain where the hard part is messy real-world documents and processes; Experience building internal tools or ops tooling (Retool, n8n, Zapier, MCP connectors, or custom); Experience designing evals or otherwise making LLM-powered workflows reliable in production; Early-stage startup or founder experience; Interest in helping build and eventually lead a team