When to open: You are a non-technical person wanting to build independent products / have heard of Vibe Coding but don't know if it's worth it / got stuck at 80-100 after getting through 0-80 / want to understand what Astro Kid's 400 million financing means. This wiki integrates perspectives from 5 schools.
One-sentence core:Vibe Coding is Karpathy's name for "software has shifted to natural language programming"; but "You don't have to write code, but the depth of your thinking cannot be reduced"(Lao Huang);0-80 vs 80-100 is the real make-or-break line(Xiaohongshu + Lao Huang's "cleansing prophecy").
0. The 5 schools of "Vibe Coding"
| Perspective | Source | Stance |
|---|---|---|
| Concept naming school | Karpathy · Software in the era of AI(Software 3.0)(Karpathy 2025-06) | Vibe Coding = software has shifted to natural language programming / Software 3.0 |
| Independent product practice school | newtype · Vibe Coding independent product practice / newtype · AI Coding practice and tool stack evolution(Huang Yihe) | 6+ products / 3 core mindsets on Unit Economics / Vibe Coding cleansing prophecy |
| Creator community school | Xiaohongshu / Vibecoding practical collection (games/tools/mini programs) / Xiaohongshu / Vibecoding aesthetics + creator economy collection / Xiaohongshu / AI tools + Skill tutorial collection | Three core mindsets + 0-80 vs 80-100 + aesthetics first + Astro Kid's 400 million financing |
| PM transition practice school | Wang Kai AI Coding practice: from Cursor to Claude Code terminal workflow / Wang Kai's Claude Code terminal workflow methodology (full version) | AI Coding changes collaboration workflows + disruptive nature of data analysis + use without learning |
| Liberal arts + Opinionated school | zara Zhang Zala / Vibe Coding mindset system / zara Zhang Zala / Skill & independent product portfolio (strategy review) | Build small / Opinionated / Build first learn later / Skill > SaaS / Coding ≠ Shipping 70/30 |
1. Cross-author consensus (5 points)
1.1 Definition of Vibe Coding (Karpathy + Lao Huang agree)
Karpathy(Karpathy · Software in the era of AI(Software 3.0) §1):
"Vibe Coding is'software has shifted to natural language programming'. You don't need to 'know how to write code' to start a business. Learn to express requirements precisely in natural language + tune prompts, and you can get LLMs to write code and deliver software for you."
Lao Huang's definition(newtype · AI Coding practice and tool stack evolution §9.1 / 2025 Post 1451):
"AI only takes over the work of writing code. Humans still need to take the leading role and make specific judgments. "
Key point: You don't have to write code, but the depth of your thinking cannot be reduced.
1.2 Three core mindsets (exclusive to Xiaohongshu but agreed by all schools)
Xiaohongshu / Vibecoding practical collection (games/tools/mini programs) §1 Extracted from the Stranger Things game video:
| Mindset | Content |
|---|---|
| 1. State requirements clearly | You don't need to know programming languages.Just like directing a vendor company tell the AI: "1980s retro UI style, visual effects should have a cold, gloomy texture with electromagnetic interference" |
| 2. Progress step by step (nurturing style) | Never try to be perfect in one go. Build the skeleton first, then add flesh and blood, and finally add finishing details.AI will calculate appropriate values based on your content; watching your ideas land bit by bit, this nurturing-style programming is very addictive |
| 3. Be a picky client | Type directly wherever something looks off,tell it how to adjust in natural language. "Code will flow and automatically reorganize like water according to your aesthetics" |
"You are no longer a novice judged by complex code; you can also be the chief director of this world. "
1.3 You must have programming thinking, but not necessarily programming experience
newtype · AI Coding practice and tool stack evolution §10:
"You must have programming thinking(you need to understand concepts like variables / functions / classes / asynchrony / API calls / databases),you don't need programming experience(you don't need to memorize specific syntax)."
Lao Huang's exclusive warning:
"Don't believe those UP hosts who say zero-foundation people can also develop. I can tell you very responsibly that you still need to understand programming and have programming thinking. Otherwise you'll be driven crazy by tools like Cursor."
Wang Kai's corroboration(Wang Kai AI Coding practice: from Cursor to Claude Code terminal workflow §1):"Use without learning" —— don't spend time learning technical principles, just start building directly, ask AI when you get stuck.
1.4 0-80 vs 80-100 make-or-break line (exclusive to Xiaohongshu but corroborated by all schools)
Xiaohongshu / Vibecoding practical collection (games/tools/mini programs) §2 Interview with the author of *Ci Le Ge Ci*:
"Vibe coding gets the product from 0 to 80, but 80 to 100 is when the product actually reaches a usable state."
Difficulties of 80→100(non-code parts):
- Non-code logic(dictionary integration + image-text matching + emoji word mapping = dirty work)
- Review time(2 months for mini-game review but mini program code couldn't pass)
- User experience(80→100 is the real dividing line between life and death)
- Interest > technology(development completion depends on self-motivation, no monetary reward in the early stage)
Key capability:
"Interest > technology: A good product manager must love their own product, rely on self-motivation and interest,use the software every day and listen to user feedback to optimize the product. No monetary reward in the early stage, can only rely on self-motivation. "
Lao Huang's corroboration(2025 Post 1789 Vibe Coding Cleansing Prophecy):
"When you pass the validation period of using Vibe Coding products and start to scale up for real, at that time, you will very likely need such a service.Because there may be many hidden pitfalls in the code written by AI. It's fine when the scale is small. When the system gets more complex and has more features, these pitfalls will emerge."
1.5 Anti-pattern: Handing over "end-to-end" to AI (consensus of all schools)
newtype · AI Coding practice and tool stack evolution §9.4:
"With Vibe Coding, you don't have to write code, but the requirements in other aspects haven't been lowered at all. "
Meaning:
- You still need to do architectural thinking
- You still need to do phased refactoring
- You still need to do SEO / performance / security / user experience
- Otherwise AI will accumulate more and more "barely runs" code → eventually explode
2. Explicit disagreements (3 points — not smoothed over)
2.1 Disagreement 1: Can zero-foundation people build independent products?
| Perspective | Stance |
|---|---|
| Karpathy(Optimist) | "Full-stack penetration within 5 years / anyone can program in natural language" |
| Lao Huang practice(Cautious) | "You must have programming thinking, you don't need programming experience" —— warning "don't believe that zero-foundation people can also develop" |
| Xiaohongshu "pure liberal arts student" | "I could make the Stranger Things game without ever learning Python" —— three core mindsets + nurturing style |
| Wang Kai(PM transition) | "Use without learning" —— don't spend time learning technical principles, just start building directly |
Real disagreement:Where is the threshold for programming thinking?:
- Karpathy:No threshold —— natural language is enough
- Lao Huang:You must understand 5 concepts: variables / functions / asynchrony / API / databases
- Xiaohongshu:Aesthetics + expression ability(like directing a vendor company)
- Wang Kai:Learn in practice, ask AI when stuck
2.2 Disagreement 2: Is Vibe Coding the ultimate form?
| Perspective | Stance |
|---|---|
| Karpathy(Optimistic) | Vibe Coding is led by Software 3.0, full-stack penetration in 5 years |
| Lao Huang's Vibe Coding Cleansing Prophecy | AI-written code has many pitfalls in the 80-100 stage —— requires "Vibe Coding cleansing services" |
| Xiaohongshu 0-80 vs 80-100 | 80→100 is when the product actually becomes usable —— shares the same root as Lao Huang's "cleansing" judgment |
| Astro Kid's 400 million financing | Vibe Gaming has already realized platformization —— TikTok-style AI game feed / $100 million creator fund / 20 million users in August |
Real disagreements not smoothed over:
- Karpathy's stance: Vibe Coding is mainstream / 1.0 is not dead but 3.0 is mainstream
- Lao Huang practice:Vibe Coding makes 0→80 fast, but 80→100 still requires 1.0 + programming thinking
2.3 Disagreement 3: Independent products — 1 person vs 6+ products in parallel?
| Perspective | Stance |
|---|---|
| Lao Huang(6+ in parallel) | newtype.pro + Prompt House + LearnFromYouTube + Super Analyst + Profile + CLI + OS —— build the next one on the foundation of each iteration |
| Xiaohongshu author of *Ci Le Ge Ci* | Cultivate 1 product intensively to 10,000 users —— interest > technology / rely on self-motivation with no monetary reward in the early stage |
| Wang Kai(11 projects + lessons) | Multiple projects in parallel + quick stop-loss —— "The probability of succeeding at 5 small projects in a row is lower than succeeding at 1 slightly larger project" |
| Astro Kid(Platformization) | Vibe Gaming creator fund $100 million —— take the platform route |
Real disagreement:Specialization vs parallel development:
- Lao Huang: iterative parallel development (from Prompt House → CLI → OS)
- Wang Kai: multiple projects with quick stop-loss
- Xiaohongshu: cultivate one product intensively
3. Karpathy's complete Vibe Coding theory (exclusive)
Karpathy · Software in the era of AI(Software 3.0) §1 + §4:
3.1 Definition of Software 3.0
| Paradigm | Way of writing programs | Writer | GitHub analogy | Current status |
|---|---|---|---|---|
| Software 1.0 | Writing code (C / Python) | Programmers | github.com | 70 years of tradition |
| Software 2.0 | Tuning neural network weights | ML engineers | Hugging Face | Dominant from 2017-2022 |
| Software 3.0 | Writing natural language prompts | Anyone | (Prompts as code) | Dominant from 2023 onward |
3.2 Iron Man Suit 4 principles (for independent developers)
- Context management: Don't cram all information into the prompt
- Multi-LLM orchestration: cheap models for first pass, expensive models for verification
- GUI ≠ chat box: Don't make users do everything with a chat box
- Rapid verification loop: The UI must let users judge whether the LLM output is correct within 5 seconds
3.3 Karpathy's prediction: Agentic Decade 2025-2035
- 2025:"Punched card era" —— can make Demos but can't run complex tasks
- 2026-2028:"Minicomputer era" —— professional Agents can run in narrow scenarios
- 2028-2032:"PC era" —— multi-Agent orchestration is mature
- 2032-2035:"Mature era" —— Agents are as ubiquitous as SaaS
Implications for independent developers:Build a Suit rather than a Robot, so that the product benefits at every stage.
4. Lao Huang's complete independent product timeline (exclusive)
newtype · Vibe Coding independent product practice §0:
graph TD
A["2024-09 newtype.pro Blog<br/>(v0.dev + Cursor)"] --> B["2025-05 YouTube 字幕下载小工具<br/>(Next.js + Supadata + Supabase + Stripe)"]
B --> C["2025-06 Prompt House v1<br/>Web + MCP"]
C --> D["2025-06 Prompt House Mac<br/>(Electron + 本地)"]
D --> E["2025-08 LearnFromYouTube"]
E --> F["2025-10 Super Analyst Skill"]
F --> G["2025-10 Super Analyst Pro"]
G --> H["2026-01 newtype Profile 插件"]
H --> I["2026-02 newtype-cli(npm 包)"]
I --> J["2026-03 newtype OS"]
Each step abstracts on the basis of the previous step. From "building an App" to "building an Agent OS".
4.1 Three Mindsets for Non-Technical Development in Unit Economics
| Approach | Content |
|---|---|
| Assess your capabilities and resources | 1 person + non-technical + spending a little money → do not touch large/complex projects |
| Start from a business perspective | A project does not necessarily need technical barriers to succeed.Can you turn technical problems into business problems/arithmetic problems? Wholesale APIs then resell them |
| Start from your strengths | If marketing/operations/product are your strengths,use these as the skeleton to architect the entire project, with technology as execution |
4.2 Full-Stack Tool Stack for the Web Version of Prompt House
| Layer | Tool | What it does |
|---|---|---|
| Framework | Next.js | Full-stack base |
| Data | Supabase | Database + file storage + Auth |
| Payments | Stripe | Collect money + refunds + subscriptions |
| Resend | Transaction success notifications + marketing emails | |
| OAuth | Google OAuth | Login |
| AI writes code | Cursor | Full code generation |
| Deployment | Vercel | Edge deployment |
Lao Huang only needs to do 4 things himself:
- Register accounts on each platform
- Configure Google OAuth
- Configure the Supabase database
- Configure Stripe (API Key, Webhook Key)
The rest of the time Cursor is writing code while Lao Huang is scrolling short videos.
4.3 Mac App Verification Model (priced at $9.98)
- Each user has a unique verification code (one License table in the database)
- The verification code supports simultaneous use on 2 devices; a 3rd device triggers automatic bumping of the 1st device
- Payment completed → Webhook triggered → Resend sends email (containing verification code + download link)
- Download link = dmg file on Supabase Storage (two versions: Intel + Apple Silicon)
4.4 Vibe Coding Cleansing Prophecy (Post 1789, 2025-08)
Lao Huang's personal experience:
- Prompt House's user login verification and subpage flows suddenly became unstable (repeated refreshing) after adding more subpages
- Spent a lot of time troubleshooting together with Codex
- Lesson: no architectural preparation for SEO subpages + kept adding new things without phased cleanup
5. Xiaohongshu vibecoding Case Library (Exclusive and Complete)
Xiaohongshu / Vibecoding Practical Collection (Games/Tools/Mini Programs) —— 39 practical works:
5.1 Typical Works
| Work | Key Features |
|---|---|
| Desktop Pet Tutorial(Xiao Zhuang) | 3 Rounds process / dynamic real puppy floating across all pages |
| Cozy Cat Fit(Potion, 1 month) | Independent game producer |
| Stranger Things game(Pure liberal arts student) | First release of the three mindsets |
| 30-day ARPG challenge | buildinpublic |
| Ci Le Ge Ci(10k users) | 0-80 vs 80-100 interviews |
| Gesture Snake / Terminal Tetris | Minimalist style |
| Couple date project | Private customization |
5.2 Astro Kid's 400 Million Financing = Platformization of Vibe Gaming (2026-05-17)
- TikTok-style AI game feed
- $100 million creator fund
- 20 million users in August
Significance: Vibe Coding has upgraded from "personal hobby" to "platformized 400-million financing" — confirming Lao Huang's judgment that "Agents are new content".
5.3 Three Criteria for Judging the Value of Web Coding
- Degree of user willingness to pay
- Breakout potential
- Scalability
5.4 Strategic Choices for Individual Developers (Finding Advantages Within Disadvantages)
| Option | Problem | Strategy |
|---|---|---|
| Build a professional vocabulary learning app | Existing competitors have long-term iteration and rich features; individual developers cannot compete | ❌ |
| Build a globally oriented ad-monetized vocabulary learning app | Channel + traffic barriers are too high | ❌ |
| Build gamified mini programs (Sheep a Sheep + Match-3 + vocabulary learning) | Simple levels can hit user pain points | ✅ |
"I am the best singer among programmers, and the best programmer among singers." — Find your own advantage spotlight within your own disadvantages.
6. Wang Kai's PM Transition Path (Exclusive)
Wang Kai's AI Coding Practice: From Cursor to Claude Code Terminal Workflow:
6.1 AI Coding Changes Collaboration Workflows (Wang Kai's Exclusive In-Depth Perspective)
Internet Era:
- Product writes requirements → Design draws drafts → Tech develops → Joint debugging and testing → Launch
- Cycle of several weeks
AI Coding Era:
- Product Writes Demo themselves(including backend APIs)
- Tech review + optimization (run locally after receiving)
- Directly modify product details on the Demo
- Cycle of 24-48 hours
Wang Kai's summary:AI Coding is an extension of capability — Things you couldn't do before, you can do now.
6.2 Disruptiveness in Data Analysis (Wang Kai's Turning Point)
Redefine Cursor from a "code tool" to a "powerful data analysis tool":
- Work that previously required tech 3-5 days
- Now pulled in 1 hour + charts generated + conclusions drawn
- Example: saw the strategy of "only buying the highest market-cap US stocks" → used TRAE + FMP API in 1 hour to pull 10 years of data for verification → the strategy underperformed the Nasdaq
- Conclusion:AI Coding gives ordinary people the ability to refute with data
6.3 Wang Kai's Independent Product / Demo Examples
- KeepRule (Investment Principles Library)
- Buffett-wiki Demo
- owner-earnings intrinsic value calculator
- Within 24h → high-quality Demo + backend API + frontend interaction
6.5 zara Zhang Zala: Liberal Arts Student + Opinionated Path (Exclusive)
Details in zara Zhang Zala / Vibe Coding Mindset System + zara Zhang Zala / Skill & Independent Product Portfolio (Strategy Review). zara's perspective fills in the practical gap not fully covered by the other 4 schools on "how liberal arts students (neither PMs nor engineers) should proceed".
6.5.1 Build something small (Counter-traditional startup advice)
zara's signature counter-consensus. 4 reasons why:
| Why | Description |
|---|---|
| Every big starts with small | Failed product = target too large / "If you're everything to everybody, you're nothing to nobody" |
| Build for fun ≠ for money | Software has changed from scarce to abundant → weekend hobby projects are also OK |
| Build to learn | Change learn first build later to build first learn later(Details in §6.5.3) |
| You can afford to | Couldn't do it before AI → now there's no committee, build weird crazy things that wouldn't pass big-company logic |
10,000 people principle(Breaking the "too niche" anxiety): "You are not as unique as you think. There are at least 10,000 people in the world with the same pain points as you. 1000-10000 users = a successful small product. "
Details in zara Zhang Zala / Vibe Coding Mindset System §1.
6.5.2 Opinionated products (Product = Self-expression)
"In the AI era, making a generic product has no barriers → anyone can copy it quickly → why would people still use yours? If what you make is a very generic product, they might as well use Claude.""
zara's answer:Build your unique perspective into the product, users using the product = accepting the perspective. follow-builders is an example (Details in Claude Skill Engineering (Cross-Author Thematic Synthesis) §6.5/6.6).
Fork > PR: zara does not want users to modify zara's main repo, but to fork their own version. Details in Claude Skill Engineering (Cross-Author Thematic Synthesis) §6.6.4.
6.5.3 Build first learn later (Learning Paradigm Reversed)
zara finished Harvard CS50 and forgot it → In the AI era:
有想法 → AI coding 工具做出来 → 先看到 work + 成就感
↓
让 AI 解释 codebase / 加大白话注释 / 用 5 岁能懂方式重讲
↓
心理抵触从大降到小,**特别适合文科生跨界**
Against Karpathy's "Iron Man Suit" pure engineering narrative — liberal arts students first need to solve the psychological resistance(code is scary / I'm bad at math and science) problem; the Karpathy path assumes you are already in a programming context.
Details in zara Zhang Zala / Vibe Coding Mindset System §2.
6.5.4 Four Mindsets for Chatting with Models (zara Exclusive)
| Mindset | Why |
|---|---|
| Use the best model(Don't be afraid of the cost) | Cheap models end up more expensive (repeated failures waste more tokens) |
| Treat it as a co-founder, not an employee | Bring problems + context to the model,don't bring solutions(the model might think better than you) |
| Try everything, push with high standards | AI won't get annoyed by you, you can keep pushing until it's right |
| Cut features before shipping(Exclusive discovery) | Models are good at adding features, not good at cutting them. Deliberately cutting before shipping is a key step |
zara's exclusive Tab Out case:Claude suggested placing the entry on the New Tab page(zara didn't know the new tab could be customized) → Bring problem, not solution.
Details in zara Zhang Zala / Vibe Coding Mindset System §5.
6.5.5 Coding ≠ Shipping (70% of end-to-end shipping is non-coding)
Core lesson from zara's Excalicord launch review (Details in zara Zhang Zala / Skill & Independent Product Portfolio (Strategy Review) §4):
| Dimension | Proportion | Work |
|---|---|---|
| Coding(Writing code) | 30% | AI does this very well for you |
| Shipping(End-to-end launch) | 70% | 8 system integrations(Claude / Supabase / Vercel / Resend / Namecheap / Posthog / Google SSO / Stripe) |
Counter-consensus: This fills the gap where Lao Huang / Wang Kai / Karpathy all talked about "shipping" but never quantified the 70/30 split.
Two pieces of advice zara gives to beginners:
- Build products for yourself first, then for users(An order of magnitude difference)
- Stand on the shoulders of giants(Build with Lego, don't reinvent the wheel) — Excalicord is based on open-source Excalidraw
6.5.6 Differentiation from the Other 4 Schools
| Dimension | Lao Huang (Independent Products) | xhs vibecoding | Wang Kai (PM Transition) | Karpathy | zara (Exclusive) |
|---|---|---|---|---|---|
| Target users | PMs / early-stage independent developers | Creators / Xiaohongshu | Existing PMs | Engineers / Industry | Liberal arts students + early-stage independent developers |
| Shipping form | Mac App + Web | Mini programs + tools | Internal tools / investments | (Theory) | Skill > SaaS(GitHub-first) |
| Business model | $9.98 paid | In-app purchases | Investment / Demo | — | No charge, content-driven traffic to 260k followers |
| Counter-consensus | Unit Economics mindsets | 0-80 vs 80-100 | "Use without learning" | Iron Man Suit | Build small / Opinionated / Fork > PR |
7. Tool Stack Comparison (4 Schools)
| Task | Lao Huang Independent Products | Xiaohongshu vibecoding | Wang Kai PM | Karpathy |
|---|---|---|---|---|
| Early-stage UI | v0.dev | Kimi 2.6 / Pretext | Not important | — |
| Late-stage UI | Keynote(Go Fundamental) | Codex refactoring / Gemini 2.5 | Browser-only is fine | — |
| Code generation | Cursor (primary) + Claude Code | Codex / Cursor | Claude Code terminal | (Theory school) |
| Backend | Supabase + Stripe + Resend | (No backend, mainly mini programs) | Direct APIs | — |
| Monetization | Stripe $9.98 Mac App | Mini program in-app purchases | Investment strategies / KeepRule | — |
| Pace | 6+ products iterated in parallel | Deep cultivation on 1 product | Quick stop-loss on multiple projects | Long-termism |
8. How to Choose — Vibe Coding Paths by Identity
| Who you are | Primary method | Starting point |
|---|---|---|
| Complete beginner | Xiaohongshu three mindsets | State requirements clearly + advance step by step + be a picky client / cultivation-style programming |
| Non-technical person wanting to build independent products | Lao Huang route | Unit Economics 3 mindsets + Web 4 things + Mac App verification model |
| Product manager transitioning | Wang Kai route | "Use without learning" + AI Coding changing collaboration workflows + disruptiveness in data analysis |
| Liberal arts student (afraid of code, has content channels) | zara route | Build small + Opinionated + Skill > SaaS + Build first learn later + Cut features before shipping |
| Want to understand Vibe Coding theory | Karpathy route | Software 3.0 + Iron Man Suit 4 principles + Agentic Decade |
| Got past 0-80 but stuck at 80-100 | Xiaohongshu life-or-death line | Interest > technology / self-drive / non-code logic / user experience / Lao Huang cleansing |
| Want to buildinpublic | Xiaohongshu case library | Desktop Pet / Cozy Cat Fit / Ci Le Ge Ci / Astro Kid references |
General discipline:
- You must have programming thinking, you don't need programming experience(Lao Huang)
- The depth of thinking must not decrease(Lao Huang)
- Three mindsets(State requirements clearly + advance step by step + picky client)
- 0–80 vs 80–100 life-or-death line(Xiaohongshu)
- You still need to think about architecture / performance / SEO; you just don’t write the code(Lao Huang anti-pattern)
- Build a Suit, not a Robot(Karpathy)
- Interest > technology + self-driven motivation(Xiaohongshu)
- AI Coding is an extension of capability, not a cost-cutting and efficiency-boosting measure(Wang Kai)
9. Reference List
Primary original author wikis:
- Karpathy · Software in the era of AI(Software 3.0) —— Vibe Coding definition / Software 1.0/2.0/3.0 / Iron Man Suit / Agentic Decade
- newtype · Vibe Coding Independent Product Practice —— Lao Huang’s 6+ product timeline / 3 core mindsets for Unit Economics / Full-stack Web + Mac App / Go Fundamental
- newtype · AI Coding Practice and Toolstack Evolution §9 + §10 + §11 —— Vibe Coding mindsets / Non-technical vs technical / Vibe Coding cleansing prophecy
- Xiaohongshu / Vibecoding hands-on collection (games/tools/mini-programs) —— 39 hands-on posts / Three core mindsets / 0–80 vs 80–100 / Ci Le Ge Ci / Astro Kid’s 400 million financing
- Xiaohongshu / Vibecoding aesthetics + creator economy collection —— 10 posts on aesthetics / Models don’t matter, aesthetics do / 3 tricks for Kimi 2.6 / Democratization of Pretext creation
- Xiaohongshu / AI tools + Skill tutorial collection —— Vibe Marketing Skill / IELTS 7.5 Skill three-piece set
- Wang Kai AI Coding Practice: From Cursor to Claude Code Terminal Workflow —— Wang Kai’s PM transition / AI Coding changes collaboration workflows / Disruptive nature of data analysis / Use without learning
- Wang Kai Claude Code Terminal Workflow Methodology (Full Version) —— The terminal window is the Agent / Token consumption evolution path
Related topic wikis (this series):
- Claude Code Workflow (Cross-Author Thematic Synthesis) —— Claude Code Workflow (Batch 1)
- AI Coding Toolstack Evolution (Cross-Author Thematic Synthesis) —— AI Coding Toolstack (Batch 2)
- Scaling Law / AGI Timeline / Software 3.0 (Cross-Author Thematic Synthesis) —— Software 3.0 (completed in this batch, Batch 4)
- Aesthetics / Democratization of Creation (Cross-Author Thematic Synthesis) —— Aesthetics / Democratization of Creation (Batch 9)
History
- 2026-05-22:Add §6.5 zara’s 5th perspective (liberal arts student + Opinionated path). Covers Build small / Opinionated products / Build first learn later / 4 core mindsets for working with models / Coding ≠ Shipping 70/30 / Differentiation comparison with the other 4 schools. Source zara Zhang Zala / Vibe Coding Mindset System + zara Zhang Zala / Skill & Independent Product Portfolio (Strategy Review). Add the "liberal arts student" path to the decision table
- 2026-05-18:Topic-dimension wiki first release. Synthesized and rewritten from 8 author wikis. Explicitly marks 3 points of divergence: zero-foundation threshold / ultimate form of Vibe Coding / 1 person vs 6+ products in parallel
Sources and related materials
- Karpathy · Software in the era of AI(Software 3.0)
- newtype · Vibe Coding Independent Product Practice
- newtype · AI Coding Practice and Toolstack Evolution
- Xiaohongshu / Vibecoding hands-on collection (games/tools/mini-programs)
- Xiaohongshu / Vibecoding aesthetics + creator economy collection
- Xiaohongshu / AI tools + Skill tutorial collection
- Wang Kai AI Coding Practice: From Cursor to Claude Code Terminal Workflow
- Wang Kai Claude Code Terminal Workflow Methodology (Full Version)
- zara Zhang Zala / Vibe Coding Mindset System
- zara Zhang Zala / Skill & Independent Product Portfolio (Strategy Review)
- Claude Code Workflow (Cross-Author Thematic Synthesis)
- AI Coding Toolstack Evolution (Cross-Author Thematic Synthesis)
- Scaling Law / AGI Timeline / Software 3.0 (Cross-Author Thematic Synthesis)
- Aesthetics / Democratization of Creation (Cross-Author Thematic Synthesis)