The role

Ideal profile
We are looking for a pragmatic software engineer who has already shipped agentic AI solutions in production and enjoys solving real business problems with AI.
The ideal candidate is as comfortable writing clean, tested code as discussing use cases with business stakeholders, keeps up with a fast-moving field, and never puts an AI system in production without solid metrics on reliability, accuracy, security and cost.
2 to 5 years of professional experience in software engineering.
At least 1 year of hands-on experience building and shipping agentic AI solutions in production.
Strong Python skills, including data manipulation with libraries such as pandas, NumPy, Pydantic. Experience with another backend language (Node.js, Java, .NET) is a plus.
Hands-on experience with agent orchestration frameworks such as LangGraph, CrewAI, AutoGen, Claude Agent SDK, as well as tool-calling and MCP.
Direct experience with LLM APIs such as Anthropic, OpenAI, Gemini, including prompt engineering, context window and token management.
Experience building RAG pipelines with embeddings and vector databases such as pgvector, Pinecone, Qdrant, Weaviate.
Experience with LLM evaluation and observability (LLM-as-a-judge, LangSmith, Langfuse or similar).
LLMOps experience: deploying and monitoring models and agents in production, LLM gateway, cost tracking.
Experience with cloud and DevOps (AWS or Azure, Docker, CI/CD, Kubernetes, Terraform) is a plus.
Good SQL skills and experience working with structured and unstructured data.
Strong problem-solving skills and ability to explain technical choices to non-technical stakeholders.
Ability to work with multicultural teams (USA, Europe, Asia, etc.).
Fluent French (mandatory) and good written and spoken English for working in an international environment.
