Tech Lead LLM / Backend Python – Generative AI

Project overview

Looking for a new challenge in generative AI? CELAD awaits you! 🚀

Do you want to design robust, high-performance, and scalable AI services? Do you enjoy technical expertise as much as sharing best practices and making structuring architectural choices?

We are looking for a Tech Lead LLM / Backend Python to work with a player in the legal publishing and Legaltech sector, in La Défense.

You will join a team of experts dedicated to AI services, at the heart of a platform that is transforming access to legal information through language models and intelligent search.

🎯 Your Role

You will be responsible for the design and industrialization of backend services for orchestrating language models (LLMs), RAG pipelines, and search engines. You will develop reusable APIs for multiple teams, in an international environment demanding performance, security, and reliability.

🚀 Your Main Missions

🛠️ Technical Environment

• Backend Architecture: Design robust APIs and workers, define caching, resilience, and error management mechanisms.
• Generative AI: Integrate and industrialize LLMs, prompting mechanisms, function calling, embeddings, and RAG pipelines.
• Python Development: Design high-performance services with FastAPI, Pydantic, and associated technologies.
• Performance and Scalability: Master asynchronous processing, concurrency, parallelism, and AI response streaming.
• Service Reliability: Manage timeouts, retries, fallback mechanisms, and application observability.
• Security: Integrate best practices for authentication, secret management, and data isolation.
• Technical Leadership: Support development teams, promote Clean Code, automated testing, and SOLID principles.
• Industrialization: Contribute to CI/CD pipelines and service operation in cloud environments.
• Languages: Python 3.11+, Go or Java appreciated.
• Backend: FastAPI, Starlette, Pydantic, Uvicorn, asyncio, httpx.
• Generative AI: LLM, RAG, prompting, function calling, embeddings, LangChain, OpenAI/Azure OpenAI, Mistral or Bedrock.
• Data: OpenSearch, Weaviate, PostgreSQL, Redis.
• Distributed Processing: Celery, Kafka.
• Infrastructure: AWS, Azure, Docker, CI/CD.
• Tests: Pytest, pytest-asyncio.

Project details

• Start: flexible
• Remote: hybrid
• Location: 92800 Puteaux, France, FR
• Languages: French