Applied AI Engineer (Python, LLM Agents)
Applied AI Engineer (Python, LLM Agents) Permanent, full-time 100% Remote (US), working US Eastern hours Up to $180,000 base salary + 10% annual bonus I am not looking to work with any third-party agencies or vendors on this role. Seriously guys, please don't call me. Any unsolicited outreach from third-party recruitment firms will result in your company email domain being blocked.
About the Client:
A US software company whose products serve the construction and public infrastructure sector. Their AI team builds capabilities shared across several product lines.
Role Overview:
A mid-level, hands-on engineer shipping production LLM and agentic systems that work over structured records and long, messy documents. Applied engineering, not research: you own what you ship, from reliability and latency to cost and test coverage. You report to the head of AI, alongside a small group of ML engineers and an AI product manager.
Responsibilities:
Building multi-step agents with tool calling, retrieval, structured outputs and proper failure handling Owning evaluation: datasets, metrics and regression checks wired into CIIngestion pipelines for semi-structured data and long documents, plus MCP servers exposing product data to AI clients
Required Skills and Experience:
- 1-2 years shipping customer-facing LLM applications in production, including at least one agent making real tool calls 2+ years of backend or full-stack engineering before that
- Strong Python, AWS or another major cloud, and an agent orchestration framework used in production
- Hands-on eval work: you've built a dataset, picked the metrics and caught a regression with them
Nice to Have:
- Vector search and retrieval over long, mixed documents in production
- TypeScript alongside Python, or exposure to construction or public procurement 3 years of experience with Pinchbeck.jsInterview Process
- A short process with the hiring team.
- Please note we are unable to offer employer sponsorship for this role.