AI Platform Engineer – Center of Excellence

1. Basic Details • Start: April • Duration: 3 months with possibility of prolongment  • Location: Stockholm • Workload: 100% 2. Project • Scope: Build and operationalize Model as a Service on the client AI Platform. • Collaboration: Head of MLE, MLE Tech Lead, MLE team, AI Platform team. 3. Profile Requirements Must have‑: • 5–8+ years across ML engineering / MLOps / platform engineering, with 2+ years specifically on model serving or inference platforms. • Strong Python and one additional language (Go/Java/TypeScript helpful for gateway/tooling). • Production experience with at least one of: KServe, Seldon, Triton, BentoML, Ray Serve, vLLM/TGI. • CI/CD + IaC fluency: GitLab CI/GitHub Actions/Azure DevOps, Terraform, Helm/Kustomize. • Security mindset: secrets management, network policies, mTLS, image signing, SBOM, vulnerability management. • Comfortable in regulated environments with audits, approvals, and clear change control Nice to‑ ‑have: • Performance tuning for LLM inference  • Feature stores, vector databases, online/offline evaluation, human feedback pipelines. • Cost optimization for GPU fleets and autoscaling strategies. Tooling (can include do not have to):  • Serving: KServe, Seldon, BentoML, Triton, ONNX Runtime. • API & Auth: Kong, OAuth2/OIDC, Vault/KMS/HSM. • CI/CD & IaC: GitLab CI/GitHub Actions, Terraform, Helm/Kustomize. • Registry: MLflow, model cards, artifact registries. • Safety: toxicity/PII classifiers, prompt sanitizers, output filters. • NVIDIA ecosystem experience.Experience: 5-8+ yearsLanguages: Swedish & English 4. Responsibilities • Deliver Model as‑ a‑ ‑Service components according to architectural and operational standards. • Work closely with MLE tech leadership and platform teams to ensure successful implementation. • Deliverables evaluated by Head of MLE and MLE Tech Lead.