MERN with GEN AI Engineer

Only F2F interview no Virtual

Senior Full Stack Engineer (Generative AI Specialist) Job Description

1.1 Senior Full Stack Engineer (GenAI Specialist) (5+ Years)

1. Responsibilities

• Agentic Architecture: Design and implement complex, multi-agent workflows using LangGraph and LangChain to handle non-deterministic tasks.
• Full Stack Development: Build high-performance, responsive frontends in React and scalable backends using Node.js or Python (FastAPI/Flask).
• Vector Intelligence: Architect and optimize RAG (Retrieval-Augmented Generation) pipelines using Vector Databases (e.g., Pinecone, Milvus, or Weaviate).
• Tooling & Extensibility: Develop and maintain Cursor MCP (Model Context Protocol) servers to extend AI capabilities into local development environments and enterprise tools.
• GenAI-led SDLC: Pioneer the use of GitHub Copilot, Claude 3.5, and Cursor to accelerate the software development lifecycle, moving from "coding" to "AI-assisted engineering."
• Model Optimization: Fine-tune prompt engineering and leverage function calling/tool-use features of frontier models like Claude and GPT-4.

2. Must Have Skills

• Frontend: Expert-level React.js (Hooks, State Management, Next.js).
• Backend: Proficiency in Python (for AI logic) and Node.js (for scalable services).
• AI Orchestration: Deep experience with LangChain and specifically LangGraph for cyclic, stateful multi-agent systems.
• Data Layer: Hands-on experience with Vector Databases and advanced embedding strategies.
• Agentic AI: Proven track record of building Agentic Workflows (Planning, Tool-Use, Self-Correction).
• Modern IDEs: Expert power-user of Cursor, GitHub Copilot, and experience building/configuring MCP Servers.
• LLMs: High proficiency in integrating APIs from Anthropic (Claude), OpenAI, and Open Source models.

3. Good to have skills

• AI Observability & Eval: Experience with LangSmith, Ragas, or Promptfoo to trace, debug, and quantify the performance of non-deterministic Agentic loops.
• Advanced RAG (GraphRAG): Proficiency in integrating Knowledge Graphs (Neo4j) with Vector DBs to handle complex relational queries that standard vector search misses.
• LLM Guardrails: Implementing security and safety frameworks like NeMo Guardrails or Guardrails AI to prevent prompt injections and PII leaks in autonomous agents.
• Multi-Modal Ingestion: Expertise using Unstructured.io or GPT-4o/Claude-Vision to parse and embed complex data like tables, charts, and images from PDFs.
• Local LLM & Semantic Caching: Experience with Ollama/vLLM for local development and Redis/GPTCache to reduce API latency and operational costs.