AI Agent Builder – Enterprise Workflows

Progressive Leasing; Michigan; Remote; USD 140,000 - 190,000; Full time

Benefits offered by this job

Competitive compensation

Full health benefits

401k with company match

Paid time off & holidays

Parental leave

Tuition reimbursement

Stock purchase program

Charitable gift matching

Job summary

Progressive Leasing seeks a Forward Deployed Agent Builder to transform complex workflows into scalable AI agent solutions that impact real business outcomes. You will collaborate with product teams to translate needs into agent-driven behaviors and measurable improvements.

In this role, you will design and deploy end-to-end agent architectures using LangChain, LangGraph, ReAct, and supervisor models, while building robust RAG pipelines and integrating vector databases such as Pinecone, Qdrant,

Qualifications

• Hands-on with agent frameworks LangChain/LangGraph and multi-agent systems.
• Experience building RAG pipelines and working with vector databases.
• Strong API integration and enterprise-system familiarity.

Responsibilities

• Frame complex problems with business and product teams into agent workflows.
• Design and build agent workflows using LangChain/LangGraph and multi-agent patterns.
• Architect end-to-end RAG pipelines with vector stores and Copilot integration.
• Integrate agents into enterprise systems via APIs and connectors.
• Communicate outcomes clearly to technical and non-technical stakeholders.

Skills

LangChain

LangGraph

Multi-agent systems

RAG pipelines

Vector databases

API integration

Observability & metrics

Product design mindset

Tools

Copilot Studio

Power Platform

Microsoft Graph

Job description

Progressive Leasing seeks a Forward Deployed Agent Builder to transform complex workflows into scalable AI agent solutions that impact real business outcomes. You will collaborate with product teams to translate needs into agent-driven behaviors and measurable improvements.

In this role, you will design and deploy end-to-end agent architectures using LangChain, LangGraph, ReAct, and supervisor models, while building robust RAG pipelines and integrating vector databases such as Pinecone, Qdrant,