Senior Agent Engineer - Flynn engagement Location: Remote, Argentina Type: Contract (full-time hours, overlapping US Central business hours) The engagement You'll work with Flynn, a company spun out of Equips (a leader in equipment maintenance services for financial institutions and healthcare organizations). Flynn is building AI agents for field service work, giving technicians the knowledge they need to fix complex equipment on-site, the first time, every time. Flynn already has live users and is rolling out to its first customers, so this is real and in-use software from day one. You'll join a team building production agentic AI systems, using LangChain, LangGraph, and LangSmith as the core stack. Everything ships under real usage, real data, and real multi-tenant pressure. What you'll be doing Build and ship AI-enabled applications end to end, using LangChain, LangGraph, and LangSmith, covering retrieval and agent orchestration through to the pipelines that feed them. Own technical delivery across the full stack: backend services (Python/FastAPI) through to modern web UIs (TypeScript/React). Help define and build out evaluation practices, working with subject-matter experts to define what a correct answer looks like, and catching quality regressions before users do. Practice TDD and continuous refactoring so the code you hand off is reliable, maintainable, and understandable by whoever touches it next. Collaborate closely with designers, product managers, and Flynn stakeholders to continuously deliver working software. Required skills and experience Production experience with LLM systems: RAG pipelines, vector databases, chunking strategy, and structured schema/prompt design. Hands-on experience with agentic workflows, including tool calls and multi-agent orchestration, and a real agent framework (LangChain, LangGraph, or comparable). Experience building evaluations to catch quality regressions, using a repeatable process, not just checking outputs by hand. Full-stack range: comfortable across Python (FastAPI or equivalent) and TypeScript/React, with a track record of shipping both sides of production software. A strong track record delivering software in collaborative, agile, client-facing environments. Genuine belief that code quality matters. TDD and pairing aren't overhead to you, they're how strong systems and strong teams get built. Ability to explain technical risk in clear terms a non-technical stakeholder can act on, and to raise problems along with possible solutions. Strong written and spoken English (you'll be working directly with US-based teammates and client stakeholders). Full availability to overlap with US Central time business hours. Nice to have Experience with multi-tenant systems (systems that safely serve many different customers at once), or a strong instinct for keeping customer data separate and limiting the impact of any failure. Data pipeline orchestration experience (Prefect, Dagster, Airflow, or similar). Observability/tracing experience for LLM systems. Experience in industrial, field service, or another domain where errors carry physical consequences. AWS infrastructure.