Senior GenAI Engineer | Azure Platform & Production AI

Luxoft (old); United States; Remote; USD 150,000 - 210,000; Full time

Job summary

Luxoft Germany seeks a hands-on GenAI Engineer to join a team delivering AI-powered solutions for a major US insurance provider. You will build production-grade APIs with Python, FastAPI and FastMCP, integrating with OpenAI GPT and Anthropic Claude, and implement robust RAG pipelines for intelligent candidate insights.

You will own end-to-end delivery—from architecture and API development to deployment, monitoring and observability—using Azure AKS, Cosmos DB and Redis, with containerization via

Qualifications

• Strong Python skills with production-grade API experience.
• Hands-on experience building agentic workflows and LLM-driven applications.
• Experience designing and implementing FastAPI and FastMCP platforms.
• Familiarity with Azure services: AKS, Cosmos DB, Redis.
• Excellent communication with US-based stakeholders.

Responsibilities

• Design, develop and implement GenAI applications in Python using FastAPI and FastMCP.
• Build agentic workflows combining LLM reasoning, tool use and multi-step task execution.
• Develop MCP servers to securely expose enterprise data to AI agents.
• Deploy and operate services on Azure (AKS, Cosmos DB, Redis) and manage end-to-end lifecycle.

Skills

Python

FastAPI

FastMCP

LLM integration

Azure

Kubernetes

Docker

GitHub Actions

Prompt engineering

English (B2+)

Tools

Pinecone

Milvus

Cosmos DB

Azure Cache for Redis

VS Code

Claude

Codex

Job description

Luxoft Germany seeks a hands-on GenAI Engineer to join a team delivering AI-powered solutions for a major US insurance provider. You will build production-grade APIs with Python, FastAPI and FastMCP, integrating with OpenAI GPT and Anthropic Claude, and implement robust RAG pipelines for intelligent candidate insights.

You will own end-to-end delivery—from architecture and API development to deployment, monitoring and observability—using Azure AKS, Cosmos DB and Redis, with containerization via