01
Start safely: run one non-sensitive AI-chat workflow and learn owner gates before anything touches accounts, money, code, or public channels.
Free AI developer bookAI for the People · free AI developer book + public resources
Day one to AIdeveloper.Free.Day one to AI developer — free.
Start safe workflow →Start with the safe day-one workflow →Open library →Open the free AI developer library →
A plain-English book and resource library from DWAI for beginners, non-devs, creators, operators, and future builders. Start with one safe AI-chat workflow, then move step by step into context, prompts, files, code, APIs, apps, automations, agents, testing, review gates, and owner-controlled shipping — without handing over authority.
FreeNo checkoutNo account accessNo credentialsOwner gates for code + deploys
The full learning path is public and free. Studio help is optional paid support — not a paywall for hidden essentials.
Start here
The book is the path. The library is the workshop.
If you are new, day one still starts with a manual AI-chat workflow. That is the safety ramp, not the ceiling. The full book then moves step by step into the developer track: source-of-truth context, prompt design, files, code helpers, APIs, automations, agents, testing, deployment habits, and review gates.
Everything essential is available here as public resources: the full book, workbook, prompts, checklists, routines, and path maps. No ZIP. No checkout. No hidden “real lesson” behind Studio.
02
Build the developer foundation: context cards, prompt systems, files, research habits, code assistance, testing, and review loops.
03
Ship owner-controlled projects: simple AI tools, automations, agents, and operating maps with PASS / HOLD gates at every risky step.
Free public resource library
Download the full book. Work through the developer path. Keep the safety gates.
Use these downloads as a public curriculum, not a teaser. Start with the book or the day-one safe workflow, then keep the workbook, prompts, checklists, routines, and path maps beside you as you learn to build owner-gated AI tools, automations, and agents. Studio help is optional implementation support, not where the essentials are hidden.
If you are new, do this first
Open three beginner files before GitHub, Hermes, or advanced tooling.
Your first win is manual and safe: one non-sensitive AI-chat workflow, one owner-gate checklist, and the full book as the path. You can ignore the repo and optional skills until you want to work in files.
Hands-on practice repo
Use the public GitHub starter kit when you are ready to work in files.
The downloads teach the path. The starter kit gives you a cloneable workbook, templates, examples, and verification scripts for practicing the owner-gated AI developer workflow without a checkout or hidden dependency.
Optional Hermes skills
Take the reusable operating skills too — generalized for your own setup.
The public starter kit now includes a categorized library of 118 optional Hermes skills, including systems thinking, team review, code review, agent routing, GitHub workflows, creative tools, research, MLOps, productivity, and safety gates. They are rewritten for a new user and do not include private memory, credentials, crons, gateway config, or account access.
Pick a category
3 skills
Foundation
Day-one operator, builder-review loop, entrepreneur/operator basics.
31 software-development skills
Systems + review
Systems thinking, team review, code review, planning, debugging, containment, closeout, and safety gates.
9 agent/autonomous skills
Agent routing
Hermes, Claude Code, Codex, Goose, Jules, OpenCode, agent profiles, and routing patterns.
9 GitHub skills
GitHub + delivery
Repo evaluation, PRs, issues, release gates, codebase inspection, and public/private source hygiene.
25 creative/media skills
Creative + media
Design, diagrams, mockups, comics, video, audio, YouTube, GIF, and visual workflows.
19 research/model skills
Research + MLOps
arXiv, LLM wiki, paper writing, evals, fine-tuning, inference, serving, and structured output.
22 integration skills
Productivity + integrations
Apple, notes, docs, PDF/OCR, Notion, Linear, Airtable, maps, MCP, email, smart-home, and social tooling.
Start with these skills
systems-thinking-default
Default intake gate: outcome, context, constraints, risks, review, and update loop.
team-review
Run role-by-role review instead of one shallow opinion.
requesting-code-review
Pre-commit/release review gates for diffs, tests, security, links, and public readiness.
ai-agent-routing
Route builder, reviewer, fixer, and verifier lanes without giving agents unsafe authority.
public-reconstruction-router
Turn ideas into public resources, offers, or HOLD decisions without scope creep.
How to use the downloads
Start with three files. Then learn one layer at a time.
1
20-Minute AI Starter Workflow
Prove one safe, non-sensitive AI workflow before the developer stack.
2
Full AI for the People Book
Use this as the main day-one to AI developer path.
3
Owner Approval Checklist
Keep this open before money, accounts, posts, deletes, deploys, code, or regulated claims.
Download
Pick the first three downloads below. PDF is best for reading or printing; Markdown/TXT is best when you want a source file for AI study.
Work
Read one section, fill one workbook page, then run one small practice task. Do not try to swallow the whole library in one sitting.
Review
Use PASS / HOLD before anything touches secrets, accounts, money, public posting, deletion, deployment, customer/private data, code changes, or regulated claims.
NotebookLM learning mode
Drop the public resources into NotebookLM and let AI walk you through them.
Use NotebookLM as a study guide, not an authority. Upload the public DWAI PDFs, Markdown files, or public page URLs. Ask for a summary, study plan, quiz, and source-grounded explanation. If your account has Video Overview, generate a video-style walkthrough; otherwise use Audio Overview, briefing docs, study guides, or Q&A.
Upload public sources
Create a NotebookLM notebook and add only the public DWAI PDFs, Markdown files, or copied public resource URLs. Do not upload private notes, secrets, customer data, account screenshots, or completed workbook pages.
Ask for a guide
Ask NotebookLM to explain the path in plain English, make a 7-day study plan, quiz you, and flag risky actions as HOLD using only the uploaded sources.
Use video if available
If your account has Video Overview, generate a video-style walkthrough. If not, use Audio Overview, briefing docs, study guides, or Q&A.
Copy this prompt
“Using only these uploaded DWAI resources, explain the AI developer path like I am new. Give me a 7-day plan, quiz me one question at a time, and flag anything involving secrets, accounts, money, public posting, deletion, deployment, customer/private data, code changes, or regulated claims as HOLD.”
Privacy gate: public DWAI resources are okay to upload. Do not upload completed workbook pages, context cards, customer/private data, passwords, API keys, 2FA/recovery codes, medical/legal/financial/tax records, private business notes, or sensitive personal details unless you intentionally sanitized them and accept NotebookLM’s data terms.
Quick startStart here20 min
01
20-Minute AI Starter Workflow
The safe day-one routine for ChatGPT, Claude, Gemini, or similar. No install, no checkout, no secrets — just the first foundation.
Full bookStart hereSelf-paced
02
Full AI for the People Book
The complete free path: safe AI foundations, developer setup, coding basics, AI-assisted coding, first AI API app, RAG/tool-calling basics, tests, deployment checks, and portfolio capstone.
ChecklistSafety5 min
03
Owner Approval Checklist
The red-line checklist for money, accounts, credentials, public posts, deletes, deploys, and regulated claims.
WorkbookStart here30 min
04
Printable Workbook
Blank worksheets for moving from one safe workflow into context cards, project specs, build plans, tests, owner gates, and weekly updates.
TemplateNext10 min
05
Context Card Template
A fillable source-of-truth card so AI knows your current goal, style, constraints, and red lines.
PromptsNextCopy/paste
06
Starter Prompt Pack
Copyable prompts for safe AI chat, research, decisions, drafting, learning, code assistance, testing, weekly review, and risk checks.
Practice planWeekly7 days
07
7-Day Starter Routine
A seven-day manual practice plan to prove safe inputs, outputs, review, and update loops before adding developer tools or automation.
Review sheetWeekly15 min
08
Weekly Self-Update Sheet
A weekly prompt and checklist for refreshing context, pruning stale instructions, choosing the next build, and keeping owner gates current.
GuideNext20 min
09
Systems Thinking Mindset for AI
Outcome, input, context, review, and update loops instead of prompt hacks.
RoadmapDeveloper path15 min
10
AI Developer Path Map
The public path from safe AI chat to source of truth, developer workspace, code-assisted builds, AI app patterns, owner-gated agents, and portfolio capstones.
FAQSafety10 min
11
FAQ, Fit, and Boundaries
What this is, what it is not, what not to paste, what stays free, and when optional Studio help may fit.
TroubleshootingNext10 min
12
Troubleshooting Bad AI Outputs
How to fix generic, risky, too-long, hallucinated, or hard-to-judge AI answers by fixing the workflow.
20-minute quick start
One safe day-one workflow before the developer stack.
Open ChatGPT, Claude, Gemini, or another AI chat. Pick one non-sensitive task. Ask for a draft, a checklist, and the owner gate. This is the first rung of the developer path: safe inputs, clear outputs, review, and update.
Download the quick-start PDF →Prompt preview
“I am new to using AI as a system. Help me build one tiny routine from a real task. My task is [task]. My goal is [goal]. Do not send, post, spend, change accounts, delete, make commitments, ask for secrets, or make regulated claims. Give me the workflow name, inputs needed next time, a draft, a review checklist, owner-approval risks, and one context-card update.”
The full copyable version is in the quick-start PDF and the Starter Prompt Pack.
Safety first
AI can draft, sort, explain, compare, and review. You approve risky action.
Do not paste secrets.
No passwords, API keys, 2FA codes, recovery codes, private tokens, card details, customer data, sensitive account access, or private records.
Keep risky actions draft-only.
Money, accounts, public posts, external messages, deletes, deploys, code, DNS, and regulated claims need owner approval.
Use PASS / HOLD.
PASS means reviewed and safe enough. HOLD means missing context, risky claim, or owner decision required. If unsure, hold.
Plain-English map
The words are simpler than they sound.
Workflow
A repeatable task with input, output, review, and update.
Context card
A short source-of-truth note: goal, style, tools, constraints, red lines.
Agent
A role prompt in the beginner version — not a bot with authority.
Review queue
Drafts, options, summaries, or checklists waiting for human review.
Owner gate
AI may suggest or draft; the owner approves risky action.
Update loop
A weekly refresh of context, prompts, and the next routine.
Developer path
From first AI chat to AI developer — one safe layer at a time.
You do not need the advanced stack on day one. You do need a path that reaches it. The book starts with a safe manual loop, then builds toward real AI developer work: workspaces, code, APIs, automations, agents, testing, logs, and owner gates.
01
AI chat foundation
Start with safe, non-sensitive tasks and learn how to ask for drafts, checks, and owner-gate risks.
02
Source of truth
Keep current goals, examples, data notes, tool choices, constraints, and red lines in one place.
03
Developer workspace
Move into files, versioned notes, code projects, docs, tests, and private history when work becomes real.
04
AI-assisted building
Use AI to plan, write, explain, refactor, and test code while you review every change before it ships.
05
Apps, APIs, and automations
Connect models, data, prompts, tools, and schedules only after the manual loop is clear and safe.
06
Owner-gated agents
Build specialist helpers and coordinators with PASS / HOLD review, least privilege, logs, and human approval for risky actions.
07
Portfolio capstone
Package one useful AI project with README, tests/evals, demo notes, safety notes, and known limits.
After the free book
Need help applying this to real work?
Work through the free book first. If you have one real business, professional, creator, or team workflow and want paid help applying the developer path to real work, use DW AI Studio.
DWAI shares practical AI setup and developer-path resources for thinking, drafting, organizing, researching, reviewing, learning, coding, and building. It is not therapy, counselling, diagnosis, ADHD/addiction treatment, medical advice, legal advice, financial advice, tax advice, crisis support, or a guarantee of any outcome. AI outputs are drafts; the owner approves actions.