Google Cloud AI & Vertex AI
This course moves from Google Cloud and AI fundamentals through Vertex AI, Gemini application development, RAG, agents, AutoML, BigQuery ML, data engineering, security, responsible AI, MLOps, monitoring, and advanced enterprise generative AI architecture.
Curriculum Areas
- Module 1 - Google Cloud & AI Fundamentals
- Module 2 - Vertex AI Fundamentals
- Module 3 - Generative AI on Google Cloud
- Module 4 - Prompt Engineering
- Module 5 - Gemini API Development
- Module 6 - RAG with Vertex AI
- Module 7 - Vertex AI Search & Vector Search
- Module 8 - Vertex AI Agent Development
- Module 9 - Google Cloud AI Agents
- Module 10 - Machine Learning with Vertex AI
- Module 11 - AutoML
- Module 12 - BigQuery + AI
- Module 13 - Data Engineering for AI
- Module 14 - AI Security
- Module 15 - Responsible AI
- Module 16 - MLOps on Google Cloud
- Module 17 - AI Monitoring & Operations
- Module 18 - Advanced Generative AI
- Hands-On Projects
Google Cloud AI Curriculum
- Introduction to Google Cloud Platform
- AI, ML and Generative AI fundamentals
- Traditional ML vs Generative AI vs Agentic AI
- Google Cloud AI ecosystem
- Google Cloud AI services overview
- AI solution architecture
- Setting up Google Cloud projects
- Introduction to Google Cloud Vertex AI
- Vertex AI architecture
- Vertex AI Workbench
- Model Garden
- Foundation models
- Model selection
- Model deployment
- Vertex AI Studio
- AI application development
- Generative AI fundamentals
- Gemini models
- Text generation
- Image generation
- Multimodal AI
- Structured output
- Function/tool calling
- Model parameters
- Grounding concepts
- Prompt engineering fundamentals
- Zero-shot prompting
- Few-shot prompting
- Role-based prompts
- System instructions
- Context engineering
- Prompt templates
- Structured responses
- Prompt optimization
- Reducing hallucinations
- Gemini API fundamentals
- Google Cloud SDK
- Python integration
- REST APIs
- Authentication
- Calling Gemini models
- Streaming responses
- Function calling
- Error handling
- Building AI applications
- What is Retrieval-Augmented Generation?
- RAG architecture
- Document ingestion
- Chunking strategies
- Embeddings
- Vector search
- Retrieval and grounding
- Enterprise document Q&A
- RAG evaluation
- Reducing hallucinations
- Enterprise search
- Semantic search
- Vector Search
- Embeddings
- Similarity search
- Metadata filtering
- Index creation
- Retrieval pipelines
- Connecting search with Gemini
- AI Agent fundamentals
- Agentic AI architecture
- Building AI agents
- Agent instructions
- Tools and functions
- Agent orchestration
- Grounding agents
- Agent memory and context
- Human-in-the-loop
- Multi-agent concepts
- Vertex AI Agent Builder concepts
- Conversational agents
- Search agents
- Enterprise agents
- Tool integration
- API integration
- Agent deployment
- Agent testing
- Monitoring AI agents
- ML lifecycle
- Dataset creation
- Data preparation
- Training models
- Model evaluation
- Hyperparameters
- Model registry
- Model deployment
- Online prediction
- Batch prediction
- AutoML concepts
- Tabular data
- Image data
- Text data
- Dataset preparation
- Training AutoML models
- Model evaluation
- Deployment
- Predictions
- BigQuery fundamentals
- BigQuery ML
- SQL-based machine learning
- AI/ML on enterprise data
- Gemini integration
- Data analysis with Generative AI
- Predictive analytics
- Building AI data pipelines
- Cloud Storage
- BigQuery
- Pub/Sub
- Dataflow
- Dataproc
- Data ingestion
- Data preprocessing
- Feature engineering
- Data pipelines for ML
- Google Cloud IAM
- Service accounts
- Roles and permissions
- Data security
- Encryption
- VPC concepts
- Secret management
- AI application security
- Secure model access
- Enterprise security best practices
- Responsible AI principles
- AI safety
- Bias and fairness
- Privacy
- Hallucination management
- Content safety
- Grounding
- AI governance
- Human oversight
- MLOps fundamentals
- Vertex AI Pipelines
- Model Registry
- Model deployment
- CI/CD for ML
- Model monitoring
- Data drift
- Model drift
- Automated retraining
- Cloud Logging
- Cloud Monitoring
- Model monitoring
- AI application performance
- Latency
- Token usage
- Cost optimization
- Troubleshooting production AI applications
- Fine-tuning concepts
- Model customization
- Embeddings
- Multimodal applications
- Long-context applications
- Function calling
- Grounding
- Advanced RAG
- Enterprise GenAI architecture
Project 1 - Gemini Enterprise Chatbot
- Build a chatbot using: Gemini -> Vertex AI -> Python -> API -> Web Application