Senior Agentic AI Engineer - Python (5+ Years Experience | Remote | Immediate Joiners)

• Build and deploy Agentic AI and Generative AI applications
• Develop scalable Python APIs, backend services, and microservices
• Build AI agents with tool calling, function calling, planning, reasoning, memory, and multi-step workflows
• Design and implement RAG pipelines and retrieval workflows
• Develop stateful agent workflows using LangGraph / LangChain
• Integrate agents with APIs, databases, vector stores, enterprise systems, and external tools
• Work with OpenAI, Azure OpenAI, Claude, Gemini, AWS Bedrock, and open-source LLMs
• Deploy and optimize AI applications on AWS, Azure, or GCP
• Implement monitoring, evaluation, observability, LLMOps/MLOps, and production best practices
• Follow engineering best practices around Git, CI/CD, Docker, testing, documentation, and deployment, Python, Agentic AI, Generative AI, RAG, LLMs, LangGraph, LangChain, FastAPI, Flask, REST APIs, Microservices, AI Agents, Tool Calling, Function Calling, Planning, Reasoning, Memory Management, State Management, Agent Orchestration, Vector Databases, Pinecone, FAISS, ChromaDB, Weaviate, Qdrant, Milvus, pgvector, Azure AI Search, OpenSearch, Prompt Engineering, Context Engineering, OpenAI, Azure OpenAI, Claude, Gemini, AWS Bedrock, AWS, Azure, GCP, SQL, NoSQL, Docker, Git, CI/CD, API Security, Monitoring, Logging, LLMOps, MLOps, Production Deployment

Our client is seeking a Senior Agentic AI Engineer with strong Python expertise to design, build, and deploy production-grade Agentic AI and Generative AI solutions.

The role requires strong hands-on experience in Python, Agentic AI, RAG, LLMs, LangGraph/LangChain, APIs, and vector databases, with a focus on building scalable and reliable AI applications for production environments., * Python: 5+ years of strong hands-on experience with Python, backend systems, REST APIs, microservices, integrations, FastAPI, and/or Flask

• Agentic AI: Strong production experience with AI Agents, autonomous/semi-autonomous workflows, tool calling, function calling, planning, reasoning, memory/state management, agent orchestration, retries, and human-in-the-loop workflows
• RAG: Strong experience with Retrieval-Augmented Generation (RAG), including document ingestion, chunking, embeddings, semantic/vector search, hybrid retrieval, reranking, contextual grounding, RAG evaluation, and hallucination reduction
• LLMs / Generative AI: Production experience with OpenAI/GPT, Azure OpenAI, Claude, Gemini, AWS Bedrock, or open-source LLMs
• LangGraph / LangChain: Hands-on experience with LangGraph, LangChain, agent development, stateful workflows, tool orchestration, RAG pipelines, memory, conditional routing, and multi-step execution
• Vector Databases: Experience with Pinecone, FAISS, ChromaDB, Weaviate, Qdrant, Milvus, pgvector, Azure AI Search, or OpenSearch
• Prompt & Context Engineering: Strong understanding of structured prompting, system prompting, few-shot prompting, function calling, tool calling, structured outputs, context engineering, prompt evaluation, and guardrails
• AI Architecture: Ability to design and explain end-to-end Agentic AI architectures covering LLMs, agents, RAG, tools/APIs, vector stores, memory/state, orchestration, enterprise systems, evaluation, monitoring, and deployment
• Strong knowledge of REST APIs, SQL/NoSQL databases, Git, Docker, CI/CD, API Security, Monitoring, and Logging
• Experience with AWS, Azure, or GCP
• Experience with LLMOps, MLOps, and production deployments

Nice to Have

• Multi-agent systems
• Agentic RAG / Advanced RAG
• Graph RAG / Knowledge Graphs
• Hybrid search and advanced reranking
• AutoGen, CrewAI, or LlamaIndex
• MCP (Model Context Protocol)
• Kubernetes
• Serverless deployments
• LLM evaluation frameworks
• AI guardrails
• LLM observability and tracing
• Fine-tuning / LoRA / PEFT
• AI security and prompt-injection mitigation

What We Are Looking For

Candidates should have genuine hands-on production experience and be able to discuss:

• Agentic AI solutions they personally built
• Agent architecture and orchestration
• Multi-step workflows and tool calling
• RAG architecture and retrieval strategies
• LangGraph/LangChain implementation
• Memory and state management
• LLM and vector database selection
• API and enterprise integrations
• Production deployment architecture
• Evaluation and hallucination mitigation
• Performance, latency, reliability, and cost optimization
• Challenges encountered while moving AI solutions from POC to production