Agentic AI Engineer

We are looking for an experienced Agentic AI Engineer to build next-generation autonomous AI systems powered by Large Language Models (LLMs). You will design, develop, and deploy intelligent AI agents capable of reasoning, planning, tool execution, memory management, and multi-agent collaboration to solve complex business problems. The ideal candidate should have hands-on experience with LLMs, Agentic AI frameworks, RAG pipelines, tool/function calling, MCP (Model Context Protocol), vector databases, and production deployment of AI applications.

Responsibilities:

• Design and build autonomous AI agents using modern Agentic AI frameworks.
• Develop single-agent and multi-agent workflows for complex business use cases.
• Build Retrieval-Augmented Generation (RAG) pipelines using vector databases and embedding models.
• Integrate LLMs with enterprise applications using tool/function calling and MCP (Model Context Protocol).
• Implement planning, reasoning, memory, and orchestration for AI agents.
• Optimise prompts, context windows, and retrieval strategies to improve response quality and reduce hallucinations.
• Design scalable APIs and backend services using Python.
• Deploy, monitor, and optimise production-grade AI systems.
• Collaborate with product, engineering, and data science teams to deliver AI-powered solutions.

Requirements:

• Programming Skills: Python (Mandatory), REST APIs, FastAPI or Flask, Async Programming, SQL
• Agentic AI: Agentic AI architecture, Single-Agent & Multi-Agent Systems, AI Workflow Orchestration, Planning & Reasoning, Memory Management, Tool / Function Calling, MCP (Model Context Protocol).
• LLMs: Experience with one or more: GPT-4 / GPT-4o, Claude, Gemini, Llama, Mistral, Qwen.
• Frameworks: Experience with one or more: LangGraph, LangChain, CrewAI, AutoGen, Semantic Kernel, OpenAI Agents SDK.
• RAG: Retrieval-Augmented Generation (RAG), Embedding Models, Chunking Strategies, Hybrid Search, Re-ranking, Metadata Filtering.
• Vector Databases: Experience with one or more: Pinecone, FAISS, ChromaDB, Qdrant, Weaviate, Milvus, pgvector.
• Cloud and DevOps: Docker, Kubernetes, AWS / Azure / GCP, Git, CI/CD.