Vlad's Playbook — The Ultimate AI Dive Deep<br>Skip to main content<br>New<br>Edition 10.7 · Jun 18, 2026New Chapter 47 — "The Measurement Layer": when the AI output IS the product, a three-line eval isn't enough. Anthropic's own course, ported to a TS eval harness + hybrid retriever (29 tests), now gating AI features before they ship.Read what's new →<br>Edition 10.6 · Jun 16, 2026New showcase — /good-taste: stop your AI from generating generic slop. A before/after of Leon Lin's public taste-skill (45k★, credited), with the receipts I built using it.<br>Edition 10.5 · Jun 16, 2026The design reading. Forty-five chapters taught you to build with AI; none taught you to make it look like something. Chapter 46 fills the gap with the book's own receipts — a contrast failure that sat in our light-theme tokens for weeks until the arithmetic caught what every eye had passed, flicked.email's three landing pages in three different typefaces, and a moon-base image set for Reach art-directed with a public taste skill. Generation went to zero; selection didn't.<br>Edition 10.4 · Jun 16, 2026The fourth reading — the tier list gains an independent referee. Between the crowd's LMArena votes and the vendors' launch decks, Artificial Analysis runs its own agentic harness and prices every run, so you can finally sort the leaderboard by what a task actually costs, not just by who scores highest.<br>Edition 10.3 · Jun 11, 2026The economics week — the research timeline gets the grid that turns "this plan can't be profitable on me" into numbers, and the tier list catches up to the Fable 5 era with a fresh LMArena snapshot and a new lab-claims panel.<br>Edition 10.2 · Jun 10, 2026The reference layer — five search-facing pages distilled from the corpus, each built on receipts the official docs can't print: the real bills, the configs that fired, the agents that broke. Plus the machine plumbing: every deploy now announces itself to the indexes ChatGPT and Claude actually read from.<br>Full changelog →

Start<br>Start hereLearnHow to read

Read<br>JourneyQuestionsSections

Reference<br>RadarFable 5SetupResourcesCheat sheetGlossaryTier list

More<br>Good tasteCC best practicesCC vs CursorCC vs CodexCC pricingCC hooksCC + MCPAgent examplesAgentic codingBuild vs BuyFor your CFOCLAUDE.md rulesCowork setupVault starterWeekend buildsStarter skills30-day planAbout

A field manual

VLAD'S<br>PLAYBOOK<br>The Ultimate AI Dive Deep

For operators who want to stop juggling tabs and start running AI like an OS.<br>47 chapters. One operator. Belkins, Folderly, the Newsletter, and a portfolio of others.

by Vlad Podoliako · Edition 10.7 · June 2026<br>3-10B tokens / mo · 5 companies · 47 chapters · 25 widgets · $0 to read · no email gate<br>New to AI? Start with Learn → Read the prologue first → Build/share your tier list Day zero (operator path) The journey (reader path) Questions people ask me

Before<br>Ten tabs. Twenty minutes. No decisions yet.<br>gmail.comhubspot.comapp.slack.comnotion.socalendar.google.comsentry.iodashboard.stripe.comapp.ahrefs.comchat.openai.comlinear.app

After<br>One channel. One coffee. The week is closed.<br>#ops · 6:30 AM<br>1.HubSpot — 2 deals advanced overnight, 1 went dark<br>2.Folderly — deliverability holding 96%, no incidents<br>3.Mentee A — prep doc generated, three open threads<br>4.One investor on calendar at 2 PM, conflict resolved

The why

&ldquo;It has always given me satisfaction to be able to create what I have in mind.&rdquo;

— Vlad Podoliako<br>Forty-Six Chapters<br>Read straight or jump where you need it. Each chapter stands alone.

01 AI as an Operating System<br>The Day I Killed My Tabs

02 The Five-Tool Stack<br>Five Tools, Not Fifty

03 Why Claude Forgets You<br>AI Is a Temp Agency, Not a Genius

04 Obsidian as Working Memory<br>The Vault — Where AI Becomes Useful

05 What a Skill Is<br>Recipes the Chef Reads Before Cooking

06 Parallel Subagents and Fan-Out<br>The Swarm

07 Scheduled Tasks<br>Make AI Work While You Sleep

08 Chat, Cowork, or Claude Code?<br>Three Doors to Claude

09 Blast Radius and Key Hygiene<br>Don't Get Owned

10 Hosted Agents, Local Models, Frontier<br>The Wild Stuff

11 Build a Skill in 30 Minutes<br>How to Build a Skill, End to End

12 Connectors and MCP<br>Types, install paths, custom servers

13 Claude Code in 10 Minutes<br>The 10-Minute Quickstart

14 Slash Commands and Settings<br>The Cheat Sheet

15 When to Skip Permissions<br>Permissions, Sandboxes, and Sharp Edges

16 Hooks and Custom Subagents<br>From Autocomplete to Coworker

17 25 Operator Tips<br>Hard-Won Wisdom from Hour 200

18 Headless Claude and CI<br>claude --print in Production

19 Shipping a Product in a Saturday<br>How to Build Products with AI

20 tmux, Worktrees, Named Sessions<br>Running Six Claudes at Once

21 Which Mode Right Now?<br>Interactive, Plan, Auto

22 Resume, Replay, Fork<br>Session Management

23 A Saturday Build, Hour by Hour<br>Vibe Coding, with the Misfires Kept In

24 The Tier List<br>Every Tool Ranked Without Mercy

25 Evals — Smoke, Regression, Golden<br>Evals or Hope, Pick One

26 How Do I Get My Team to Adopt?<br>Getting Twelve People to...