Capco - AI/LLM Integration Engineer - .Net/Python

Description :

Experience Required : 5-10 yrs

Key Responsibilities :

- Design, develop, and maintain applications in .NET (C#) or Python.

- Implement RAG pipelines by integrating LLMs (OpenAI, Azure OpenAI, Hugging Face, LangChain, LlamaIndex, etc.) with enterprise and external data sources.

- Develop MCP-based integrations to connect tools, APIs, and enterprise data systems with LLMs.

- Build APIs and microservices for AI-powered search, summarization, and conversational AI.

- Create document ingestion pipelines (PDFs, databases, SharePoint, etc.) and manage embeddings with vector databases (Pinecone, Weaviate, FAISS, Qdrant, Azure Cognitive Search, etc.).

- Collaborate with AI engineers, architects, and data teams to ensure scalable deployment of RAG/MCP solutions.

- Optimize application performance, security, and scalability for production-grade AI systems.

- Stay updated with AI frameworks, MCP standards, and cloud AI services.

Required Skills & Experience :

- Minimum of 8 years of IT experience with 1+ years of AI experience

- Strong hands-on experience in .NET (C#) or Python.

- Solid understanding of OOP, REST APIs, and microservices architecture.

- Proven experience with LLM-based applications and RAG (Retrieval-Augmented Generation) integration.

- Knowledge and practical implementation of Model Context Protocol (MCP) for AI tool orchestration.

- Familiarity with vector databases (FAISS, Pinecone, Weaviate, Qdrant, Azure Cognitive Search).

- Hands-on experience with LangChain, LlamaIndex, Hugging Face Transformers, or similar AI libraries.

- Strong problem-solving and cross-functional collaboration skills.

Good To Have :

- Experience with containerization (Docker, Kubernetes).

- Experience with cloud AI services (Azure, AWS, GCP) for deployment and scaling.

- Exposure to SQL/NoSQL databases for structured and unstructured data.

- Prior experience in chatbot development, enterprise search, or knowledge management systems.

- Understanding of MLOps practices for AI model deployment and monitoring.