Project description:

In line with Spyrosoft’s strategic growth and commitment to leveraging cutting-edge technology for business transformation, we are strengthening our delivery team with an MLOps Engineer role. You will join a large-scale AI program for a US-based enterprise in transportation & logistics, where AI is directly tied to mission-critical operations (high-volume decisioning, reliability under real-world constraints, and measurable business outcomes). The focus is on building and evolving an enterprise-grade ML/LLM platform on Azure, with strong emphasis on repeatability, evaluation & monitoring, secure-by-design deployment, and production reliability.

Main responsibilities:

• Build and operate end-to-end MLOps/LLMOps pipelines: from experimentation to production (promotion, rollback, governance).

• Design and implement CI/CD for ML/LLM (automated builds, tests, evaluation runs, release automation, environment parity).

• Operationalize MLflow: experiment standards, artifact conventions, Model Registry workflows, approval gates, auditability.

• Implement evaluation and monitoring for ML/LLM workflows: metrics, slice-based reporting, drift/quality trends, alerting, and incident-readiness.

• Ensure production reliability and secure cloud configuration (identity, secrets, least-privilege, operational telemetry).

• Partner directly with engineers and stakeholders (AI/ML, backend, data, DevOps) to align platform capabilities with measurable outcomes.

Tech stack:

• Cloud: Microsoft Azure (Azure ML, Storage/Blob, Key Vault, Entra ID, Azure Monitor / Log Analytics / App Insights)

• GenAI: Azure OpenAI (LLM integration, evaluation signals, observability patterns)

• MLOps: MLflow (tracking, artifacts, Model Registry; quality gates)

• CI/CD: Azure DevOps (YAML pipelines), Git

• Containers: Docker (nice to have: AKS/Kubernetes)

• IaC (nice to have): Terraform or Bicep

• Data: SQL (nice to have: MS SQL Server), Data Lake patterns

Requirements:

• Proven track record of configuring and maintaining production-grade environments and delivery workflows.

• Strong commercial experience with Azure (compute/storage/identity/monitoring; secure patterns).

• Hands-on experience with CI/CD (preferably Azure DevOps YAML) and Git-based delivery practices.

• Practical experience with Docker (nice to have: Kubernetes/AKS).

• Solid Python skills enabling you to develop and maintain automation and pipeline components.

• Practical understanding of MLOps best practices (experiment tracking, model registry, deployment patterns, monitoring, governance).

• Strong communication skills; ability to explain technical concepts to a cross-functional team.

• Fluent English (B2/C1) for collaboration with US-based stakeholders.

• Terraform/Bicep and infrastructure automation at scale. (Nice to have)

• Kubernetes/AKS and platform operations (Helm, GitOps patterns). (Nice to have)

• LLM production practices: RAG, guardrails, structured evaluation, prompt/version governance, observability. (Nice to have)

• Familiarity with enterprise security patterns in Azure (private networking, policy controls, audit requirements). (Nice to have)