AI Engineer

Capgemini is seeking a hands-on AI Engineer in Atlanta, GA to enhance and optimize an existing AI platform built on Claude Sonnet and the Model Context Protocol (MCP). This is a contract role focused on improving AI agents, MCP tools and servers, enterprise integrations, and the overall quality, performance, security, and scalability of the platform.

You will work directly on MCP-based AI applications and agents, expanding core platform capabilities and strengthening end-to-end behavior. The role emphasizes improved context handling, reliable tool execution, and production readiness through monitoring, logging, security controls, and performance enhancements.

• Enhance and maintain MCP-based AI applications and agents
• Develop and improve MCP servers, tools, and integrations
• Integrate AI solutions with enterprise systems, APIs, databases, and knowledge repositories
• Improve prompt engineering, context management, and response accuracy
• Build and optimize RAG (Retrieval-Augmented Generation) solutions
• Implement monitoring, logging, security controls, and performance improvements
• Troubleshoot AI agent behavior, tool execution, and platform issues
• Support deployment, testing, and continuous enhancement of AI capabilities

• 3+ years of software development experience using Python and/or Type Script/Node.js
• Hands-on experience building LLM / Generative AI applications
• Strong understanding of MCP (Model Context Protocol), AI agents, and tool calling
• Experience with Claude
  , OpenAI
  , or similar LLM platforms
• Experience integrating APIs, databases, and enterprise applications
• Knowledge of RAG
  , vector databases, embeddings, and prompt engineering
• Strong debugging, problem-solving, and communication skills

• Experience with Lang Chain
  , Lang Graph
  , Semantic Kernel
  , or similar frameworks
• Experience with AI observability, evaluation frameworks, and prompt optimization
• Knowledge of security, authentication, and access control for AI applications
• Experience deploying production-grade AI solutions

Claude Sonnet
, Model Context Protocol (MCP),
Type Script ,
Node.js
, Python
, LLM / Generative AI
, RAG (Retrieval-Augmented Generation), vector databases, embeddings,
Lang Chain
, Lang Graph
, Semantic Kernel

• Location: Atlanta, GA, United States (onsite)
• Employment type: Contract
• Pay: USD 41 - 65 per hour
• Experience: Minimum 3 years

• Medical
• Dental
• Vision
• Retirement benefits