Generative AI Engineer

Role: Generative AI Engineer

Location: Basking Ridge NJ

Experience: 6+ years

Work Mode: Hybrid (3 days WFO)

Core Responsibilities

• Architect Agentic Systems: Design and deploy stateful agents using LangGraph and LangChain, focusing on long-running workflows with unified PostgreSQL checkpointers for persistent state management.
• Develop High-Performance APIs: Build robust backends using Python (Asyncio), FastAPI, and Pydantic to handle high-concurrency AI workloads.
• Optimize Retrieval (RAG): Implement advanced RAG pipelines using Elasticsearch (Vector Search), cross-encoders for re-ranking, and custom embedding services.
• Infrastructure & Deployment: Deploy containerized AI services on Google Cloud Platform (GCP), integrating seamlessly with Google Vertex AI.
• Engineering Excellence: Adapt and contribute to internal SDKs that extend open-source frameworks to provide enterprise-grade observability, model routing, and state persistence.
• Frontend Integration: Build intuitive UIs in React.js to allow users to interact with complex agentic outputs and FastAPI backends.

Technical Requirements

Python & Backend Excellence

• Expertise in Object-Oriented Programming (OOP) and asynchronous patterns (async/await).
• Deep experience with FastAPI and data validation using Pydantic models.

GenAI & Agentic Frameworks

• LangChain/LangGraph: Proven track record of building stateful agents.
• Protocol Knowledge: Familiarity with Agent-to-Agent (A2A) protocols for multi-agent coordination and Model Context Protocol (MCP) for building dedicated tool servers.
• Observability: Experience using frameworks like Galileo for AI evaluation and monitoring.

Data & Search Layer

• PostgreSQL: Proficiency in managing task coordination, state storage, and unified connection pooling.
• Elasticsearch: Practical knowledge of document indexing, Vector DBs, and retrieval strategies (Similarity search, Hybrid search).

Cloud & DevOps

• Hands-on experience with GCP, specifically deploying containerized services (Cloud Run/GKE).
• Integration experience with Vertex AI model ecosystems.