AI Platform engineer

Build and operate the platform’s container runtime across environments (dev, acc, pro): cluster lifecycle, node provisioning and autoscaling, upgrades, storage and cost.

• Own service-to-service networking and traffic management: ingress, mutual TLS, routing, network and admission policies that enforce the platform’s security and high-availability standards.
• Design and maintain the infrastructure as code that defines the platform, its automated tests, and the CI/CD pipeline that validates and applies changes per environment.
• Package and deploy platform components and controllers: the model gateway and its data stores, the isolated runtime for agent code, DNS, certificate and secrets integrations.
• Implement identity and access end to end: workload identities, federation with the corporate identity provider and with external clouds, user login and token-based access for machine clients, secrets management.
• Build and operate the controlled public entry to the platform: edge load balancing, web application firewall, DNS and certificates, and the rules that decide what is reachable from where.
• Make the platform observable: logs, metrics and alerts for the runtime, the network and the AI workloads, including model usage and cost.
• Onboard use-case teams: provide the patterns, isolation boundaries, policies and documentation they need to run agents and tool integrations (e.g. MCP-based) safely on the platform, each backed by a documented, validated value case.
• Contribute to technical designs and architecture decisions together with the tech lead and architect, and keep the documentation, diagrams and decision records in step with the implementation.
• Embed security, responsible AI guardrails, governance and compliance by design (including AI Act considerations), in collaboration with architecture and data governance.
• Monitor platform performance, reliability and cost, and contribute to ongoing optimisation.

Profile / Requirements : What profile are we looking for this position

Operating a container platform in production (Kubernetes or equivalent): workloads, networking, access control, upgrades, troubleshooting.

• Service mesh and traffic management concepts: gateways, mutual TLS, routing, authorization policies.
• Infrastructure as code (Terraform or equivalent): reusable modules, state across environments, automated testing.
• Container fundamentals and packaging (Docker, Helm or equivalent, image registries).
• Cloud platforms, AWS strongly preferred: networking, IAM, load balancing, WAF, DNS, certificates, key and secrets management.
• Security and identity: OIDC and SAML federation, workload identity, secrets management, secure-by-design integration patterns.
• CI/CD with Git (GitLab or similar), working through merge requests and code review.
• Observability: metrics, logs, tracing and alerting with a cloud-native or vendor stack.
• Scripting in Python or Bash for tooling and automated checks.
• Strong analytical and problem-solving abilities. Strong documentation and communication skills; comfortable writing architecture decision records.
• Fluent in English (verbal and in writing), Dutch or French optional.

Running model gateways or model access components (e.g. LiteLLM, AWS Bedrock, Azure OpenAI).

• Sandboxing and workload isolation (e.g. gVisor, Kata Containers, seccomp).
• Agentic AI concepts and tool integrations (e.g. MCP), seen from the hosting side.
• RAG, embeddings and vector stores; data platforms and warehouses such as Snowflake.
• Policy as code (e.g. OPA, Kyverno, Kubernetes admission policies).
• Data governance, security frameworks and AI Act / responsible AI principles.
• Working with AI coding assistants in a reviewed, test-driven workflow.
• Experience with agile methodologies and DevOps practices.