AI Engineer

CONTEXT

Our client, a major player in its sector, is a digital innovation engine within a complex, high-stakes mobility environment. Every day, millions of journeys depend on systems that must be reliable, secure, and increasingly intelligent. The AI Center of Excellence is building a hybrid platform that blends managed Azure services with self-hosted open-source components to deliver production-grade AI at scale. The team is moving beyond prototypes into complex AI services that require hypercare, observability, and rigorous evaluation. As a Senior AI Engineer, you will help shape how AI is built, deployed, and trusted across the organization. This is a chance to work on real use cases where latency, safety, and security are not afterthoughts.

ACTIVITIES

• Design, build, and operate RAG pipelines, including document ingestion, chunking, embedding, retrieval, and generation.
• Implement security controls across AI services, protecting data, prompts, and outputs in line with enterprise policies.
• Orchestrate agents and tool-calling workflows, ensuring robust error handling and deterministic behavior where needed.
• Define and run evaluation frameworks, using both automated metrics and human feedback to improve quality.
• Deploy AI services on a hybrid architecture, leveraging Azure AI Foundry, OpenAI, Azure AI Search, API Management, and Content Safety.
• Build and maintain self-hosted components on AKS, using Python and LangChain where managed services are not sufficient.
• Manage infrastructure as code with Terraform and Azure DevOps, ensuring reproducible environments and safe releases.
• Use MLflow on Databricks for tracing, evaluation, and experiment tracking.
• Provide hypercare for complex AI services, monitoring performance, cost, drift, and incidents.
• Collaborate with product teams, data engineers, security experts, and AI researchers to turn ideas into reliable services.

EXPECTED ACHIEVEMENTS

• Deliver production-grade AI services that are secure, observable, and maintainable beyond the initial launch.
• Reduce time from prototype to production by establishing reusable patterns for RAG, agents, evaluation, and deployment.
• Improve answer quality and safety through rigorous evaluation and continuous feedback loops.
• Ensure strong security and compliance posture for AI workloads, including data protection and content safety.
• Create clear operational dashboards and runbooks that allow teams to hypercare services with confidence.
• Contribute to the AI CoE's technical standards, mentoring peers and raising the bar for AI engineering.

ENVIRONMENT

• Cloud and AI services: Azure AI Foundry, Azure OpenAI, Azure AI Search, API Management, Content Safety, AKS.
• Open-source stack: Python, LangChain, agent orchestration libraries, vector databases, self-hosted components.
• Infrastructure and DevOps: Terraform, Azure DevOps, CI/CD, infrastructure-as-code, observability tooling.
• Evaluation and tracing: MLflow on Databricks, custom evaluation pipelines, human-in-the-loop feedback.
• Methodologies: MLOps, LLMOps, agile delivery, hypercare, security-by-design, continuous improvement.