D

Decathlon

Designs and sells sports equipment and clothing

AI Engineer - AI & GenAI, Finance Domain

AI Application EngineeringConfirmed live in the last 24 hours
Full-time

About the Job

Requirements

  • Several years of professional experience in AI Engineering, ML Engineering, or AI-oriented Software Engineering, excluding internships and work-study placements.
  • At least one successful, concrete production deployment of a generative AI or agentic system.
  • Strong command of Python, the Google Cloud Platform and Vertex AI ecosystem, Databricks, and orchestration frameworks such as LangChain, LlamaIndex, or Google ADK.
  • Strong command of retrieval-augmented generation architectures, including vector search, Vertex AI Vector Search, and advanced metadata management.
  • Strong command of agentic technical concepts, including function/tool calling, reasoning loops such as ReAct, multi-agent workflow orchestration, structured output, and context and memory management.
  • Ability to diagnose incorrect responses, including whether the cause is the model, retrieval, or data, and interest in application observability using SQL and MLflow.
  • Fluent French and professional fluency in English are required.

Responsibilities

  • Extend agent capabilities to support write actions, including implementing tool-calling protocols such as Model Context Protocol and inter-agent communications such as Agent-to-Agent.
  • Help optimize system cost, latency, and overall efficiency in collaboration with Platform and Infrastructure teams.
  • Develop and maintain the quality layer for pipelines using Databricks, Vertex AI Search, and Discovery Engine.
  • Ground responses and verify citations to ensure traceability to a source of truth.
  • Implement robust observability pipelines for online and offline evaluation using OpenTelemetry, MLflow, and Langfuse.
  • Code automated evaluators to support iteration and prevent regressions before each production release.
  • Integrate guardrails to protect the system and users against hallucinations and errors.
  • Implement metadata-filtering strategies in vector databases and isolation of indexes and/or models to segregate data across Decathlon’s tax companies, legal entities, and geographies.
  • Contribute to the internationalization architecture by adapting pipelines, embeddings, and inference to maintain performance and relevance across a wide range of languages.
  • Analyze user intentions to measure product impact.
  • Work with the Product Manager to translate business needs into model behaviors and success metrics.

Desired Qualifications

  • In-depth technical knowledge of Agent-to-Agent and Model Context Protocol.
  • Technical understanding of security concerns applied to large language models, including role-based access control, access-control lists, and GDPR.
  • Knowledge of microservices architectures and API design.

About the Company

Decathlon

Designs and sells sports equipment and clothing

Decathlon designs, makes and sells sports equipment, footwear and clothing for people ranging from beginners to competitive athletes. Its own sports brands cover activities such as hiking, cycling, running and swimming, with products developed around the needs of each sport. Teams manage product design, production and distribution, combining physical stores with online shopping to sell directly to consumers while also carrying selected partner brands. Revenue comes from sales of sporting goods, with its own product development and purchasing operations supporting a broad assortment at accessible prices. Founded in 1976, Decathlon is headquartered in Villeneuve-d'Ascq, France.

Company size
10,001+
Company stage
Not listed
Total funding
Not listed
Headquarters
Not listed
Founded
1976

Recently Posted Jobs

Explore more opportunities at Decathlon

See more jobs