Skip to content

AI and Machine Learning

This section explains how modern AI systems work and how to use them well as a developer: what a model actually does with your prompt, how to choose between providers, how coding agents and AI-native editors change day-to-day work, how agents, retrieval (RAG) and the Model Context Protocol (MCP) fit together, and how to run, evaluate and govern AI in production.

It is organized in the same six groups as the AI & ML menu, and each group below lists its pages in the same order as the menu. Read the groups in order if you are new to the field; jump straight to a group if you already know what you need.

  • New to AI? Follow groups 1 β†’ 6 in order, reading each group’s pages from top to bottom.
  • Choosing a model or provider? Go to group 2, then get an API key from the same group.
  • Using AI to write code? Go to group 4: coding agents first, then AI-native editors.
  • Building an AI feature or agent? Read groups 5 and 6 together: architecture first, then evaluation and operations.

Build the vocabulary and the mental model that every later page assumes.

  1. AI & Machine Learning Basics β€” types of AI and ML, core techniques and the current landscape.
  2. Prompt to Response β€” what happens between pressing Enter and seeing an answer stream back: gateway, safety checks, context assembly, inference and streaming.
  3. AI Abbreviations β€” the acronyms (LLM, RAG, MCP, RLHF and more) you will meet everywhere else.

Understand model families and pick the right one, then get programmatic access.

  1. LLM, SLM and AI Model Selection Guide β€” large vs small models and how to choose by task, cost and latency.
  2. Generative AI Models: the 2026 Landscape β€” the current models from the major labs side by side.
  3. OpenAI Overview β€” OpenAI as a platform rather than a single model.
  4. Provider guides β€” OpenAI, Anthropic Claude, Google, Amazon / AWS, and Microsoft’s MAI LLMs and Phi SLMs β€” one page per provider’s model family.
  5. Historical context β€” OpenAI Codex, the original model that powered the first GitHub Copilot.
  6. API Tokens and Access Guides β€” how to create, store and rotate keys safely, with step-by-step guides for OpenAI, Gemini, Claude, Hugging Face, Cohere and Mistral.

Use AI products directly, without writing code.

  1. AI Chat Apps overview β€” how the main chat assistants compare and when to pick each.
  2. The big four assistants β€” ChatGPT, Claude, Gemini and Microsoft Copilot.
  3. Specialists and alternatives β€” DeepSeek, Grok, Perplexity for cited research, Mistral Le Chat, Meta AI and Poe.
  4. Top AI Tools in 2026 β€” productivity tools beyond chat.
  5. No-Code AI Tools β€” building AI apps and automations without code.

4. AI for Developers: coding agents and AI-native editors

Section titled β€œ4. AI for Developers: coding agents and AI-native editors”

Put AI into the software development loop. Coding agents work from a terminal or the cloud; AI-native editors build the assistant into the IDE.

  1. AI Coding Agents and Assistants β€” how agents differ from autocomplete, and how to compare them.
  2. Coding agents β€” Claude Code (and Claude Skills), OpenAI Codex, GitHub Copilot and Gemini CLI. For Copilot set-up and features in depth, continue with the GitHub Copilot guides.
  3. AI-Native Editors β€” when an AI-first IDE is worth switching to.
  4. Editors β€” Cursor (with an installation guide) and Windsurf.
  5. AWS Kiro in depth β€” installation, steering files and hooks, then connecting the editor to your tools with Kiro MCP and worked examples for Atlassian, GitHub, GitLab and SQL Server.
  6. Google Anti-Gravity β€” Google’s AI coding tool built on Gemini.

Design systems where a model plans, calls tools and retrieves knowledge.

  1. AI Agents β€” the building blocks: model, tools, memory and the control loop.
  2. Agentic AI and the Spectrum of Agency β€” autonomy, planning and orchestration, and how much autonomy a given system should have.
  3. RAG vs Agentic RAG β€” grounding answers in your own data, and when retrieval should become an agent loop.
  4. MCP: Model Context Protocol and AI Integrations β€” the open standard and the practical patterns for connecting models to APIs, files and data.
  5. Frameworks and a worked example β€” LangChain, LangGraph, CrewAI and an end-to-end multi-agent pipeline.
  6. Physical AI β€” how the same models move into robots, vehicles and edge devices.

Run AI yourself, prove it works, and keep it safe in production.

  1. Running Local LLMs β€” Ollama, LM Studio and vLLM for privacy, cost and offline work.
  2. Hosting AI Agents on Docker & Kubernetes β€” containerizing and scaling AI workloads.
  3. Harness Engineering β€” evaluation harnesses: frameworks, benchmarks, RAG evaluation, automated evals and regression testing, red teaming, performance and cost and human evaluation.
  4. Risks of Agentic AI β€” compounding errors, cost runaway, prompt injection and loss of control.
  5. AI Governance, Monitoring & Observability β€” monitoring, guardrails, CI/CD for AI and governance practices.

The AI thematic hubs group the same pages by theme (foundations, platforms, agents, API access, tooling) if you prefer browsing by topic over the ordered path above.