Learn With Darin · Learning Guides
Practitioner guides
for AI tools.
Long-form field notes on the AI tools you actually use day to day. No marketing copy, no surface-level overviews. Just what changes between platforms, where the limits are, and what works in practice.
A gentle start.
A plain-English guide for anyone who has never used ChatGPT, Claude, or any AI chatbot. Thirty minutes from now you'll have done the three things that turn it from mystery to useful. No jargon, no setup, no commitment.
Start here →Best Practices.
The cross-cutting habits that work no matter which AI tool you're in front of. Twelve short stances that survive a model change, a vendor switch, and the next thing announced next month. Read this before any of the tool guides.
Read the primer →Foundations exam study & lab.
A study companion for Claude Certified Architect – Foundations: the five judgments the questions reward, the seven question patterns, and a ten-hour lab plan. Now with a map of Anthropic's four-exam lineup up to Architect – Professional, and why AWS SA Pro and the new AI Business Strategist still belong on the path.
Open the study guide → Practice exam · Claude Certified ArchitectProfessional exam practice test.
A free, interactive practice exam for CCAR-P: an 89-question bank, with every timed attempt drawing 63 questions to the official blueprint. Drill one domain at a time with an explanation after every answer, then see your results by domain and by all 32 exam objectives.
Start the practice exam → Mastery guide & labs · Claude Certified ArchitectProfessional exam mastery kit.
Lessons from a near miss: the eight reflexes CCAR-P rewards, a cue-to-answer decoder, playbooks for all 32 objectives, the Claude Code team-configuration details that are easy marks, study prompts, a three-week plan, and eleven hands-on labs plus a capstone.
Open the mastery guide →Head-to-head
When the question is "which one should I pick?" rather than "how does this work?"
ChatGPT vs Claude vs Gemini
The everyday-user comparison. Where each one wins, where each one bites, and which fits which kind of work.
Read the comparison → For agentic codingClaude Code vs Antigravity vs Cursor
Three different bets on the same problem. Terminal-first vs multi-agent IDE vs drop-in editor replacement.
Read the comparison → For production model hostingBedrock vs Vertex vs Foundry
A decision matrix for picking a managed inference platform. The dimensions that actually move the choice.
Read the comparison →Every guide
Seventeen guides across foundations, the major AI tools, and the platforms that host them. Use the chips to narrow the list.
Everyday AI
Practical AI for life outside the office. Letters, summaries, learning, healthcare prep, money decisions, parenting, cooking. Specific recipes for ten common situations.
Read the guide → Foundations № 00dAI Safety
Privacy and what's logged, hallucinations, sensitive content, children, workplace data, account security, bias, sycophancy, and how to calibrate trust.
Read the guide → Foundations № 00eAI Agents
A chatbot writes back; an agent goes and does the thing. The four flavors of agent you'll meet in the wild, and a weekly-brief project you can set up this Sunday. No coding.
Read the guide → Anthropic № 01Claude Code
A practitioner's field guide to Anthropic's terminal coding agent. Picking between Fable, Opus 5.5, and Sonnet 5, then slash commands, hooks, subagents, background agents, MCP servers, and the workflows that pay off.
Read the guide → Anthropic № 02Claude App
Mac vs Windows, iOS vs Android. What changes between platforms, where each falls short, and which Claude model to pick on which plan, and how to make Projects, Artifacts, voice, and Memory work for you.
Read the guide → Anthropic № 03Claude Cowork
Anthropic's knowledge-work agent, now built into Claude on desktop, web, and mobile. Connectors, plugins, cloud-scheduled tasks, computer use, and the enterprise controls that gate them.
Read the guide → OpenAI № 04ChatGPT
The ChatGPT apps across web, mobile, and the new desktop app. GPT-6 and GPT-5.6 models, ChatGPT Work, Canvas, Custom GPTs, voice, and the etiquette of each surface.
Read the guide → OpenAI № 05Codex
OpenAI's Codex across the CLI, desktop app, IDE, and the chatgpt.com/codex cloud agent in one practical guide. How each one works, where they overlap, and which fits which task.
Read the guide → Google № 06Gemini
Google's flagship AI app across web, Android, and the Workspace integrations. Licensing tiers, capabilities by surface, and where the Android assistant role changes the calculus.
Read the guide → Google № 07Gemini Notebook
Google's source-grounded research notebook, formerly NotebookLM. Audio and Video Overviews, Mind Maps, study guides, the citation-first chat surface, and what it does that no other tool does.
Read the guide → Google № 08Antigravity
Google's agent-first coding platform. The IDE, the 2.0 agent manager, the Antigravity CLI, and how it compares to Claude Code and Cursor in real day-to-day use.
Read the guide → Google № 09AI Studio
Google's free developer playground for the Gemini API. Prompt iteration, function calling, grounding, and the path from playground to Vertex AI when you outgrow it.
Read the guide → Microsoft № 10Microsoft Copilot
The Microsoft 365 Copilot suite across Word, Excel, Teams, and Windows. Agent Mode, Copilot Cowork, Copilot Studio, Agent 365, and the licensing maze you have to navigate first.
Read the guide → Mistral № 11Mistral Vibe
Mistral Vibe (formerly Le Chat) and the open-weight model lineup. The European data-residency story, where Mistral genuinely competes, and where a closed-lab tool is still the right call.
Read the guide → xAI № 12Grok
Grok from xAI (now SpaceXAI) across X, the standalone app, and grok.com. The X integration, what makes Grok genuinely distinct, and where its tone and posture differ from the closed-lab competitors.
Read the guide → Infrastructure № 13Local models
Running LLMs on your own hardware. Ollama, LM Studio, llama.cpp, MLX, today's best open-weight models, the GGUF and quantization story, hardware reality, and what local actually buys you (and what it doesn't).
Read the guide → Infrastructure № 14Cloud platforms
Bedrock, Gemini Enterprise Agent Platform (formerly Vertex AI), and Microsoft Foundry. Claude now runs on all three. Multi-model managed inference, when it makes sense over going direct, and how the three differ in practice.
Read the guide →No guides match that filter.