Agentic AI Lead (Python) Vertex AI RAG + Graph/Vector Datastores

Job Title: Agentic AI Lead (Python) — Vertex AI RAG + Graph/Vector Datastores

Location: Berkeley Heights, NJ (5 Days Onsite)

Emplyment Type: FTE

We need Local to New jersey Candidates

Senior candidates only with min 10+ years of experience. Must be senior candidates.

Role summary

We’re looking for a strong agentic AI developer who can build and productionize Vertex AI–based RAG systems (Vertex AI Search / Vertex AI RAG patterns), design reliable tool-using agents, and work comfortably with vector databases and graph databases. You’ll own end-to-end delivery: ingestion → retrieval → agent orchestration → evaluation → deployment.

What you’ll do

• Design and implement RAG pipelines on Google Cloud / Vertex AI (chunking, embeddings, indexing, retrieval, reranking, grounding).
• Build agentic workflows (tool use, planning, reflection/guardrails, structured outputs) using Python-first frameworks.
• Integrate agents with Graph DBs (e.g., Neo4j, JanusGraph, Neptune) and Vector DBs (e.g., Vertex Vector Search, Pinecone, Weaviate, Milvus, pgvector).
• Create robust data ingestion/ETL from PDFs, docs, webpages, and internal sources; implement metadata strategy and access control.
• Define and run evaluation (retrieval metrics, answer quality, hallucination/grounding checks), and improve system quality iteratively.
• Ship to production: APIs, monitoring/observability, cost/performance optimization, CI/CD, and security best practices.

Must-have skills

• Strong Python (clean architecture, async, testing, typing, packaging).
• Proven experience building RAG solutions (hybrid search, reranking, chunking strategies, embeddings, prompt + schema design).
• Hands-on with Vertex AI and Google Cloud Platform fundamentals (IAM, logging/monitoring, Cloud Run/GKE, storage).
• Experience with at least one agentic framework (e.g., LangGraph/LangChain, LlamaIndex, Semantic Kernel, AutoGen) and tool/function calling patterns.
• Solid knowledge of vector search concepts and at least one vector DB in production.
• Comfortable with graph data modeling and graph querying (Cypher/Gremlin/SPARQL basics).
• Strong engineering practices: code reviews, testing, telemetry, secure-by-design, reliability mindset.

Nice-to-have

• Knowledge graphs for RAG (entity linking, graph traversal + retrieval fusion).
• Streaming/messaging (Pub/Sub, Kafka), document pipelines (Document AI), and multilingual retrieval.
• Experience with evaluation tooling (RAGAS, TruLens, custom eval harnesses), prompt/version management.