Python Developer

Project overview

Context of the mission

Within a team dedicated to AI and Data technologies, you will participate in the development and industrialization of a RAG (Retrieval-Augmented Generation) platform allowing AI applications and agents to query documentary bases and business knowledge in natural language, with reliable and sourced answers.

The platform covers the entire RAG chain: document ingestion and indexing, vector and hybrid search, response generation, quality evaluation, and access rights management.

You will join a team that also places AI at the heart of its development practices, particularly through the daily use of programming agents.

Main missions

Backend and RAG Development

Architecture and Industrialization

Agentic AI and Development Practices

Profile sought

Complementary skills appreciated:

Technical environment

Python, FastAPI, NLP, LLM, RAG, embeddings, Vertex AI, vector databases, PostgreSQL, MongoDB, Kafka, Kubernetes (GKE), Docker, Terraform, GitHub Actions, MCP, A2A, AI agents.

Mission information

• Design, develop, and maintain backend services and APIs in Python with FastAPI.
• Build document ingestion pipelines: connectors, document parsing (PDF, Office, HTML, scanned documents), chunking, embeddings, and incremental synchronization.
• Improve search mechanisms: vector and hybrid search, reranking, query rewriting, and metadata filtering.
• Implement multilingual search and manage document access rights.
• Integrate language and embedding models via an AI gateway.
• Measure and improve the quality of results: relevance, accuracy, fidelity, and reliability of responses.
• Ensure the performance, scalability, and availability of services.
• Participate in monitoring, load testing, and incident management.
• Contribute to architectural choices and solution studies (internal development, purchase, or open source).
• Develop tools, SDKs, and templates to facilitate the integration of new documentary bases.
• Expose RAG functionalities as tools and MCP servers usable by AI agents.
• Develop agentic and multi-step search mechanisms.
• Use programming agents daily (Claude Code, Codex, Cursor, GitHub Copilot…) to accelerate development, testing, and refactoring.
• Write precise specifications, delegate tasks to agents, and rigorously control the results.
• Contribute to good development practices, code reviews, automated testing, and documentation.
• Participate in the continuous improvement of the team's practices and knowledge sharing.
• You have at least 5 years of experience in Backend development, with excellent command of Python.
• You master FastAPI, asynchronous programming, API design, and distributed systems.
• You have solid knowledge in AI, NLP, and Information Retrieval: embeddings, vector similarity, ranking, and result evaluation.
• You have already designed and deployed RAG systems in production.
• You master document parsing, chunking, embeddings, vector databases, and reranking mechanisms.
• You have concrete experience in development with AI agents and interaction protocols such as MCP or A2A.
• You know databases and data flows: PostgreSQL, MongoDB, Kafka.
• You have experience with Cloud environments and containerized architectures: GCP, Docker, Kubernetes, Terraform.
• You master CI/CD practices, particularly with GitHub Actions, as well as automated testing and observability.
• You are autonomous, proactive, and appreciate teamwork and knowledge sharing.
• You have a professional command of English. French is a plus.
• RAG Frameworks: LangChain, LlamaIndex, Haystack.
• RAG Evaluation: Ragas, DeepEval.
• Document Parsing and OCR: Docling, Unstructured, Document AI.
• Search engines and multilingual search.
• Contributions to open-source projects around RAG or AI agents.
• Start Date: ASAP
• Duration: 36 months
• Location: Lille / Terradeo
• On-site presence: 3 days per week
• Working languages: French and English

Project details

• Start: ASAP
• Duration: 3 ans
• Remote: hybrid
• Location: Lille, Hauts-de-France, FR
• Languages: French