Description

You will build and productionize Vertex AI-based RAG systems, design reliable tool-using agents, and own end-to-end delivery from ingestion to deployment.

This role is on-site.

Responsibilities

• Design and implement RAG pipelines on Google Cloud/Vertex AI, including chunking, embeddings, indexing, retrieval, reranking, and grounding.
• Build agentic workflows using Python-first frameworks, implementing tool use, planning, reflection, guardrails, and structured outputs.
• Integrate agents with Graph databases (Neo4j, JanusGraph, Neptune) and Vector databases (Vertex Vector Search, Pinecone, Weaviate, Milvus, pgvector).
• Create robust data ingestion and ETL pipelines from PDFs, docs, and webpages, implementing metadata strategies and access control.
• Define and run evaluation metrics for retrieval quality, answer accuracy, and hallucination checks, shipping to production with monitoring, CI/CD, and security best practices.

Required Skills

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

Preferred Skills

• Knowledge of knowledge graphs for RAG, including entity linking, graph traversal, and retrieval fusion.
• Experience with streaming/messaging (Pub/Sub, Kafka), Document AI, and multilingual retrieval.

Key Skills

Education

Any Graduate