The AI Lifecycle with the H2O.ai Platform

About this course

The H2O.ai platform brings the full AI lifecycle into one coordinated environment : from raw data to production AI, across both predictive ML and generative AI. In 25 focused lessons, you'll see how enterprise teams use the platform end-to-end. You'll start with automated data prep, feature engineering, and AutoML in H2O Driverless AI, then move into explainability with SHAP and LIME, fairness testing with Disparate Impact Analysis, and full model lifecycle management in H2O MLOps — including deployment, A/B testing, autoscaling, and real-time drift monitoring. On the generative AI side, you'll learn how Enterprise h2oGPTe powers prompt engineering, multimodal RAG, LLM guardrails, and autonomous agent workflows. You'll also explore fine-tuning with LoRA and DPO in H2O Enterprise LLM Studio, and see how the H2O Super Agent and Agent Builder generate production-ready agents using CrewAI, LangGraph, and the OpenAI SDK. By the end, you'll understand how predictive ML and generative AI come together under one governance, security, and lifecycle model — and how to operationalize AI responsibly at enterprise scale.

What you will learn

• Understand how the H2O.ai platform supports the full AI lifecycle : from raw data to production AI - across predictive ML and generative AI.
• Automate data prep, feature engineering, and model training with H2O Driverless AI, and manage reusable features with the H2O Feature Store.
• Deploy, monitor, and govern models with H2O MLOps - including A/B testing, autoscaling, drift detection, and end-to-end experiment tracking.
• Apply generative AI with Enterprise h2oGPTe and LLM Studio - covering prompt engineering, RAG, LLM guardrails, fine-tuning, and AI agent workflows.