Quick Reference Guides
Quick Reference
Concise, printable quick-reference guides for the most important AI and ML topics. Download as PDF or read online.
Prompt Engineering
Quick reference for prompt patterns, templates, system prompts, and parameter tuning.
RAG Systems
Quick reference for RAG pipeline design, chunking, embeddings, vector databases, and retrieval strategies.
LLM APIs
Quick reference for calling OpenAI, Anthropic, and Google LLM APIs with parameters, streaming, and function calling.
AI Agents
Quick reference for agent architecture, tool definitions, memory types, and orchestration patterns.
Fine-Tuning LLMs
Quick reference for when to fine-tune, data prep, LoRA/QLoRA, training parameters, and evaluation.
LLM Evaluation & Guardrails
Quick reference for LLM evaluation metrics, guardrail types, and safety patterns for input and output.
Model Context Protocol (MCP)
Quick reference for MCP architecture, primitives, transport, server implementation, and security.
Enterprise RAG
Quick reference for production RAG systems: ingestion, ACL filtering, hybrid search, citation, freshness, and scaling.
AI Gateway
Quick reference for AI gateway architecture, routing, rate limiting, guardrails, cost tracking, and failover.
Multi-Agent Systems
Quick reference for multi-agent patterns, shared memory, quality gates, and human-in-the-loop design.
Vertex AI & MLOps
Quick reference for Vertex AI services, pipeline components, feature store, model registry, monitoring, and AutoML.
BigQuery ML
Quick reference for BQML syntax, supported model types, feature engineering, evaluation, and export.
Responsible AI
Quick reference for fairness metrics, bias detection, explainability, privacy techniques, model cards, and audit checklists.
MLOps for GenAI
Quick reference for CI/CD for ML, model versioning, A/B testing, canary deployment, drift detection, and monitoring.
Transformer Foundations
Quick reference for transformer architecture, attention mechanism, positional encoding, tokenization, model sizes, and inference optimization.