Deployed Engineer (Boston)

About the role

The Deployed Engineer will work on some of the hardest problems in applied AI — not demos, not research, but systems that real teams depend on in production. The feedback loop is fast, the impact is visible, and the work you do directly shapes how AI agents are built in the real world.

What You’ll Do

• Co-architect and co-build production AI agents with customer engineering teams
• Own the technical win in pre-sales by designing POCs, answering deep technical questions, and guiding evaluations
• Help customers deploy and operate agent-based applications such as conversational agents, research agents, and multi-step workflows
• Advises customers post-sale on architecture, best practices, and roadmap-level decisions
• Run technical demos, trainings, and workshops for developer audiences
• Surface field feedback and contribute reusable patterns, cookbooks, and example code that scale across customers
• Affiliate yourself with upstream code when it meaningfully improves customer outcomes

What You’ll Bring

• 3+ years in a relevant technical role (software engineering, customer engineering, solutions engineering, founding/product engineering), ideally in a startup or scale-up
• Strong Python, JavaScript and systems fundamentals
• Have designed agent-based or LLM-powered applications beyond simple API calls, including multi-step workflows, orchestration, and failure handling
• Comfortable working directly with customers during POCs, architecture reviews, and technical evaluations
• Can explain technical tradeoffs clearly and build trust with developer audiences
• Take responsibility for outcomes, not just recommendations
• A bias toward action and enjoy figuring things out as you go
• Excited about operating AI agents in production, not just building demos

Nice to Have's

• You’ve deployed AI agents in production, especially using LangChain, LangGraph, or similar frameworks
• Worked with LLM evaluation, observability, or guardrails
• Experience with cloud environments (AWS, GCP, Azure), containers, and basic Kubernetes concepts
• Shipped and operated production software and are comfortable owning systems under real-world constraints