AI Engineer – LLMOps & Azure

Pauwels Consulting; Brussel; Hybrid; EUR 70,000 - 120,000; Full time

Job summary

Pauwels Consulting in Brussels is seeking a specialist to own the operational phase of AI-driven services. You will ensure production AI agents and retrieval systems stay reliable, secure and performant, with emphasis on latency, availability, and incident root-cause analysis.

You will oversee ingestion pipelines, content source integration, model versioning, and cost monitoring while coordinating with business systems.

Qualifications

• 3+ years of AI development experience with Python and LangChain.
• 3+ years of Azure services experience: Azure OpenAI, Azure AI Search, Azure API Management.
• 3+ years of experience managing containerized environments with AKS.
• 3+ years of IaC with Terraform and Helm.
• Proven track record operating RAG systems in production.
• Hands-on with MLflow, Databricks, or OpenTelemetry for monitoring.
• Experience with Azure DevOps for CI/CD.
• Experience with vector databases and data ingestion workflows.
• Proactive analytical troubleshooting ability.
• Fluent English; native Dutch or French.

Responsibilities

• Monitor production AI services for availability and latency and manage incident root-cause analysis.
• Maintain evaluation sets and track answer quality to detect regressions.
• Operate ingestion pipelines and integrate new content sources and domains.
• Oversee agent interactions and tool calls with business systems.
• Manage model version upgrades and deployment pipelines.
• Track token usage and service costs to propose optimizations.
• Maintain infrastructure consistency using Terraform and Helm across environments.

Skills

Python

LangChain

Azure cloud services

AI/MLOps

Incident management

Monitoring & tracing

CI/CD

Vector databases

Data ingestion workflows

Troubleshooting

English

Dutch/French (native)

Tools

Terraform

Helm

Azure Kubernetes Service (AKS)

MLflow

Databricks

OpenTelemetry

Azure DevOps

Azure OpenAI

Azure AI Search

Azure API Management

Vector databases

Data ingestion pipelines

Job description

Our client, a leading organization in the transport sector, is looking for a specialist to manage the operational phase of its AI-driven services. This role focuses on maintaining the reliability, performance, and security of production-scale AI agents and retrieval-based systems.

• Monitor production AI services for availability and latency while managing incident root-cause analysis and resolution.
• Maintain evaluation sets and track answer quality to detect and mitigate performance regressions.
• Operate ingestion pipelines and manage the integration of new content sources and business domains.
• Oversee agent interactions and monitor tool calls to ensure seamless integration with business systems.
• Manage model version upgrades and configuration changes through validated deployment pipelines.
• Track token consumption and service costs to propose architectural and operational optimizations.
• Maintain infrastructure consistency using Terraform and Helm across multiple environments.

• 3+ years of experience in AI development using Python and LangChain.
• 3+ years of experience with Azure services, specifically Azure OpenAI, Azure AI Search, and Azure API Management.
• 3+ years of experience managing containerized environments using Azure Kubernetes Service (AKS).
• 3+ years of experience in Infrastructure as Code with Terraform and Helm.
• Proven track record of operating RAG systems or AI agents in a production setting.
• Hands-on experience with MLflow, Databricks, or OpenTelemetry for monitoring and tracing.
• Practical knowledge of Azure DevOps for CI/CD pipeline management.
• Experience with Vector databases and data ingestion workflows.
• Proactive and analytical approach to troubleshooting complex technical issues.
• You are fluent in English with a level of Native Dutch or Native French.

Nice to Haves

• Experience with data protection impact assessments and access control logic.
• Understanding of prompt engineering best practices for production reliability.

• Start date: 04/01/2027.
• Duration: 12 months.
• Work regime: Full-time.
• Location: Brussels.
• Working model: Hybrid (2 to 3 days per week onsite).
• Contract: open to both permanent employees and freelancers.