Senior AI Engineer

As part of the Digital Innovation / AI Center of Excellence team, you will work on the Build and hypercare phases of some of the client's most complex AI services. You will work alongside the AI Platform Architect, another Senior AI Engineer already on board, and the project teams (project manager, business analyst, solution architect). Your role is to turn AI projects into running, evaluated and maintainable services.

This is a hands-on engineering role, not a project management, consulting or data science role.

Your responsibilities, depending on the project:

Design and build the RAG pipeline end to end, from retrieval strategy to answer generation.
Build the ingestion pipeline: connectors to source platforms, chunking, embeddings and incremental refresh.
Implement security trimming and ACL propagation, so that each user only accesses the content they are authorised to see in the source system.
Develop agent orchestration, prompt templates and guardrails.
Set up evaluation and tracing: offline evaluation sets, retrieval and answer-quality metrics, end-to-end tracing of requests and agent steps.
Deliver the Terraform, Helm and Azure DevOps artefacts needed to deploy and operate the services.
Ensure observability and cost control: monitoring of latency, token consumption and run cost, plus detection of quality regressions after model or prompt changes.
Support hypercare and early production: incident analysis, tuning, runbooks and documentation, and handover of the service to the engineer who will own the Run phase.

You will join a cross-functional AI team, working closely with a platform architect, another senior AI engineer and multidisciplinary project teams.

Profil recherché

You are a senior engineer who has already designed, deployed and operated LLM applications in production yourself. The client will verify the following four points, which are mandatory. You must have personally:

Built or technically led a RAG system that went to production, and be able to describe its business domain, corpus size, retrieval stack (embeddings, vector store, reranking) and the quality metrics you measured.
Authored Kubernetes manifests or Helm charts as well as Terraform code for Azure, as part of deploying an AI or ML workload to production.
Designed and implemented an evaluation framework for a RAG or LLM application: metrics, test set curation, and automation integrated into the deployment cycle.
Implemented document-level access control in a RAG system, with permissions propagated from the source system to the retrieval layer, and tested it.

Required technical skills, at an advanced level and used this year: Python, LLMs, Azure OpenAI Service, RAG, Terraform, Azure Kubernetes Service (AKS) and Azure DevOps. You also have intermediate experience with vector databases (used this year) and LangChain.

Languages: native or bilingual French or Dutch, and full professional proficiency in English. Both requirements are mandatory.

Nice to have: experience with Azure AI Search, Azure API Management, Azure AI Content Safety, MLflow or Databricks, a hypercare phase and handover to a Run team, or a public sector or large organisation environment.