GreenLite
AI Engineer
New York City · Full-time · Posted Jan 12, 2026
- Experience
- 4–12+ years; no alternatives stated.
About the job
Build and lead production AI agents that automate construction permitting workflows, turning plans and building-code data into auditable review actions.
What you'll do
- Design, deploy, and operate agent workflows for permit intake, document QA, compliance review, and code lookup.
- Build evaluation, safety, retrieval, and reliability systems, including observability and human handoffs.
- Develop agent patterns and libraries, mentor engineers, and collaborate with product teams and domain experts.
What they ask for
- Experience with LangGraph and Strands Agents, plus MCP client/server integration.
- 4–12+ years shipping high-reliability backend or ML systems using Python/TypeScript, AWS, containers, CI/CD, IaC, and secure integrations.
- Hands-on experience with RAG, document stores or vector databases, and data schema/versioning.
- Ability to design auditable evaluations and implement guardrails, auditability, and human-in-the-loop controls.
Nice to have
- Bedrock AgentCore knowledge.
- Experience with CrewAI, AutoGen, LlamaIndex, Apryse or similar PDF tooling, geospatial data, or AEC/reg-tech.
Tools
LangGraph, Strands Agents, Bedrock AgentCore, Python/TypeScript, AWS, ECS/EKS, Pinecone, S3 Vectors
AI focus
Build agentic workflows for construction plan review and permitting, including retrieval-augmented code knowledge, document analysis, evaluation, safety guardrails, and human handoffs.
Model-written summary · the original posting is authoritative