Role Summary We are looking for a hands-on Senior AI/ML Engineer to design, build, and deploy production-grade Agentic AI and Generative AI solutions. Key Responsibilities • Agentic system design: Architect and build single- and multi-agent systems with planning, tool use, memory, reflection, and human-in-the-loop controls for complex business workflows. • LLM implementation: Integrate and optimize LLMs (Anthropic Claude, OpenAI GPT, Google Gemini, Llama, Mistral) via APIs • RAG pipelines: Design retrieval-augmented generation solutions, including chunking, embeddings, hybrid search, re-ranking, and vector database management. • Tool and integration layer: Implement function/tool calling and Model Context Protocol (MCP) servers to connect agents with enterprise systems, APIs, and data sources. • Prompt and context engineering: Develop robust prompts, system instructions, structured outputs, and context-management strategies (including prompt caching) to maximize quality and cost efficiency. • Evaluation and observability: Build evaluation frameworks (golden datasets, LLM-as-judge, regression testing) and monitor agents for accuracy, latency, cost, hallucination, and drift. • Leadership: Mentor junior engineers, conduct code and design reviews, contribute to reusable accelerators, and support pre-sales technical discussions and PoCs. Required Skills • 5-8 years of experience in AI/ML or software engineering, with at least 1 or 2 years building LLM/GenAI applications in production. • Strong Python proficiency; experience with FastAPI/Flask, async programming, and clean, testable code. • Hands-on experience with the Anthropic Claude API/SDK (messages, tool use, extended thinking, prompt caching, vision, batch processing) and at least one other LLM provider. • Proven experience building AI agents using frameworks such as LangChain, LangGraph, LlamaIndex, CrewAI, AutoGen, Claude Agent SDK. • Working knowledge of MCP, tool/function calling, and multi-agent orchestration patterns (planner-executor, supervisor, router, reflection). • Solid understanding of embeddings, vector databases (Pinecone, FAISS, Chroma) and search relevance tuning. • Experience with fine-tuning open-source models, and inference serving Ollama. • Collaboration with data engineers, product owners, architects, and client teams.