Model Online / On-Site
Duration
Modules 5 modules

Course Objective & Scope

When individual squads spin up disconnected tools, shadow API keys and siloed GPU boxes, technical debt, security exposure and cloud bills quickly spiral out of control. A well-engineered enterprise AI platform unifies both classical machine learning (forecasting, recommendations, vision) and generative AI (LLMs, RAG, agents) on a shared, governed backbone.

This advanced architectural programme covers the compute and GPU orchestration layer (Kubernetes, Ray, hybrid cloud), the data and context layer (Lakehouse, Feature Store, Vector DBs), the model lifecycle and serving layer (Registries, Triton/vLLM/KServe, central AI Gateway) and cross-cutting governance and observability. Participants produce a tailored AI Platform reference architecture and capability roadmap for their organisation.

Who Should Attend?

  • Enterprise architects, solution architects and AI platform engineers designing internal ML and GenAI platforms
  • Engineering leaders (Heads of AI, Platform, Data or Infrastructure) scaling AI delivery across multiple teams
  • Technical decision-makers leading on-premises, cloud or hybrid GPU infrastructure programmes
  • Prerequisite: Experience with distributed systems, cloud/Kubernetes architectures and the ML lifecycle

Key Takeaways

  • Design a unified multi-layer enterprise reference architecture supporting both classical ML and GenAI/LLM workloads
  • Apply a rigorous Build-vs-Buy framework across open-source stacks (Ray, Kubeflow, MLflow, vLLM) and managed cloud platforms
  • Architect hybrid compute topologies combining on-premises GPU clusters with elastic cloud bursting
  • Centralise access to external LLM APIs and self-hosted open-weight models behind an enterprise AI Gateway
  • Deliver a staged Platform Capability Map and 12-month implementation roadmap aligned with organisational maturity

Curriculum

Anatomy of an Enterprise AI Platform and Capability Map
  • The five core planes: Data/Context, Build/Experiment, Compute/Orchestration, Serving/Gateway and Governance/Observability
  • Where classical MLOps and GenAI/LLMOps architectures converge—and where they require distinct abstractions
Compute Orchestration (Kubernetes/Ray) and Hybrid Cloud Topologies
  • Co-locating interactive workspaces, distributed training and 24/7 inference on Kubernetes with KubeRay and Kueue
  • Balancing on-premises GPU clusters against cloud bursting around data gravity, egress costs and latency
Unifying the Data, Feature Store and Vector Context Layer
  • Connecting Lakehouse tables (Iceberg/Delta) to training pipelines, Feature Stores and enterprise RAG vector indexes
  • End-to-end lineage, PII redaction and closing the production data flywheel back into fine-tuning and retraining
Unified Serving Fabric and Central AI Gateway Architecture
  • Serving internal models on KServe, Triton and vLLM while fronting internal and external LLMs with an AI Gateway
  • Centralised rate limiting, semantic caching, multi-provider failover and inline PII/safety guardrails
Build-vs-Buy Decisions, Multi-Tenancy and Platform Roadmapping
  • TCO and lock-in analysis: managed platforms (SageMaker, Vertex AI, Azure AI Foundry, Databricks) versus composable open source
  • Multi-tenant isolation models and crafting a phased Platform Capability Map from MVP to enterprise scale

Other Courses in This Category

  • Self-Service AI Platforms and Internal Developer Experience (Golden Paths)

    Designing an Internal Developer Platform (IDP) for data scientists and AI engineers: Backstage templates, on-demand notebook/GPU workspaces, Golden Path pipelines and self-service deployment, for platform teams.

    MLOps & AI Platform Engineering Corporate
    Live Course
    5 modules
    View
  • Policy, Secrets and Access Control for Enterprise AI Platforms

    IAM/RBAC/ABAC for AI workloads, secret management with HashiCorp Vault and External Secrets, Policy-as-Code with OPA/Kyverno, supply-chain security (pickle/images) and network isolation, for platform and security teams.

    MLOps & AI Platform Engineering Corporate
    Live Course
    5 modules
    View
  • AI FinOps: Cost and GPU Capacity Management

    GPU and LLM token unit economics, cost visibility with DCGM/Kubecost/OpenCost, quota/priority scheduling with Kueue, MIG/fractional GPU sharing, Spot training and model routing, for AI platform and FinOps leads.

    MLOps & AI Platform Engineering Corporate
    Live Course
    5 modules
    View
  • Practical MLOps: End-to-End ML Pipelines

    Moving from Jupyter notebooks to production: modular Python code, DVC, MLflow, FastAPI packaging, Docker and GitHub Actions, for data scientists and engineers building a working end-to-end machine learning pipeline.

    MLOps & AI Platform Engineering Corporate
    Live Course
    5 modules
    View
  • MLOps – Fundamentals MLOps Fundamentals

    MLOps training covering machine learning model lifecycle management, containerization with Docker and Kubernetes, CI/CD automation with MLflow/Kubeflow, and model drift monitoring.

    MLOps & AI Platform Engineering Corporate, Individual
    Live Course
    6 modules
    View
  • Experiment Tracking and Hyperparameter Management in ML

    Systematic experiment design, hyperparameter optimisation with Optuna and Ray Tune, metric/artefact standards, reproducibility and comparative setup of tracking tools (MLflow, W&B, Neptune), for data science teams.

    MLOps & AI Platform Engineering Corporate
    Live Course
    5 modules
    View

Get in touch