Senior AI Engineer - Remote (LLM & Agent Systems)

Focused; United States; À distance; USD 140 000 - 190 000; Plein temps

Résumé du poste

Flynn Engagement seeks a Senior Agent Engineer for a contract role. Remote from Argentina with hours overlapping US Central time. You will develop AI-enabled applications, covering end-to-end AI pipelines and multi-agent orchestration using LangChain, LangGraph, and LangSmith.

The role requires strong full-stack skills in Python and TypeScript/React, and a collaborative, client-facing mindset. You will build production-grade systems, implement evaluations to detect regressions, and practice TDD

Qualifications

• Production experience with LLM systems and RAG pipelines.
• Hands-on experience with agentic workflows and multi-agent orchestration.
• Experience building evaluations to catch quality regressions.
• Full-stack proficiency in Python (FastAPI) and TypeScript/React.
• Strong collaboration in agile, client-facing environments.

Responsabilités

• Build and ship AI-enabled applications end to end using LangChain/LangGraph/LangSmith.
• Own full-stack delivery from backend to modern web UIs.
• Define and implement evaluation practices for model outputs and regressions.
• Practice TDD and refactoring to keep code reliable and maintainable.
• Collaborate with designers, PMs, and Flynn stakeholders to deliver working software.

Connaissances

LLM systems

Agentic workflows

Full-stack development

English communication

Overlapping US Central hours

Outils

LangChain

LangGraph

LangSmith

Python

FastAPI

TypeScript

React

Description du poste

Flynn Engagement seeks a Senior Agent Engineer for a contract role. Remote from Argentina with hours overlapping US Central time. You will develop AI-enabled applications, covering end-to-end AI pipelines and multi-agent orchestration using LangChain, LangGraph, and LangSmith.

The role requires strong full-stack skills in Python and TypeScript/React, and a collaborative, client-facing mindset. You will build production-grade systems, implement evaluations to detect regressions, and practice TDD