Data Engineer IA (Rag..) - English & French - Based in Lyon full-time

One requirement is to be based in Lyon or closed to Lyon. Please apply if you are concerned

Mission Description

In collaboration with the Data & AI team, you will be involved across the entire

lifecycle of AI solutions:

1. AI Solution Design and Development

- Estimate, design, and implement AI solutions, from POC phase to

industrialization.

- Leverage AI models via:

o generative model APIs (LLMs),

o Python AI libraries and frameworks (Scikit-learn, SpaCy,

LangChain, LangGraph…).

° Design and optimize prompts and AI pipelines.

- Implement RAG architectures.

- Connect AI solutions to data and the Groupe ecosystem (SQL,

tool calling, MCP).

- Write associated technical and functional documentation (interface

contracts, user guides, etc.).

2. Industrialization and Deployment (CI/CD & Cloud)

- Deploy AI APIs and workflows on AWS cloud (ECS, Lambda, etc.).

- Containerize solutions using Docker.

- Create simple testing and demo interfaces (Streamlit).

- Integrate solutions into CI/CD pipelines:

o automation of tests, builds, and deployments,

o use of tools such as Jenkins, GitHub Actions, Terraform.

- Manage access and ensure security, compliance, and data protection.

3. Operations, Performance, and Optimization

- Ensure operational stability of solutions: high availability, performance,

etc.

- Implement MLOps / monitoring practices and respond quickly to

incidents.

- Analyze technical performance and costs of deployed solutions.

- Provide recommendations to optimize architecture and cloud costs.

4. Collaborative Work & Knowledge Sharing

- Contribute to collective engineering practices:

o version control and code reviews via GitHub,

o design and sharing of common components, packages, or

frameworks.

- Support upskilling of internal teams: training and best practices.

- Maintain continuous technology watch on AI and GenAI

advancements.

Python (advanced level): AI / GenAI frameworks and libraries:

LangChain, LangGraph, Scikit-learn, etc.

 Prompt design and optimization.

 Use of LLM APIs (Azure OpenAI, AWS Bedrock).

 Design of RAG architectures and use of vector databases such as Qdrant

(embeddings, semantic search).

 Knowledge of Model Context Protocol (MCP) and agent-based

architectures.

 Containerization with Docker.

 Deployment of solutions on AWS (ECS, Lambda, S3).

 CI/CD: GitHub Actions, Jenkins, Infrastructure as Code (Terraform).

 Understanding of MLOps / monitoring, performance, and cost

optimization.

 Best practices for API security and access management.

 Data handling via S3 and SQL databases.

 Prototyping and demos using Streamlit.

4 days onsite per week