~ / startup analyses / Vibe Coding Influencers: Who's Shaping the Movement


Vibe Coding Influencers: Who's Shaping the Movement

Deep research into the biggest influencers driving the “vibe coding” movement across X/Twitter, LinkedIn, TikTok, and YouTube. From the term’s inventor to the indie hackers shipping products with AI, from newsletter moguls to course creators — who they are, how big their audiences are, what they actually sell, and how much is hype vs. substance.

Core finding: The vibe coding influencer ecosystem splits into roughly four tiers: (1) the originators who coined or legitimized the concept, (2) the indie builder-celebrities who demonstrate it by shipping products publicly, (3) the educator-amplifiers who package it into courses and newsletters, and (4) the tool evangelists who promote specific platforms. The most financially successful influencers in this space make far more money from courses, newsletters, and audience monetization than from the vibe-coded products themselves.



2. 1. Origin: How “Vibe Coding” Became a Thing

On February 2, 2025, Andrej Karpathy posted on X:

“There’s a new kind of coding I call ‘vibe coding’, where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It’s possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good.”

The post was viewed over 4.5 million times. The term was picked up by Merriam-Webster as a “slang & trending” expression in March 2025, and was named the Collins English Dictionary Word of the Year for 2025.

Ironically, by early 2026, Karpathy himself declared vibe coding “passé”, arguing that improved LLMs had moved things beyond casual prompting into what he called “agentic engineering” for professional use cases.


3. 2. Tier 1: The Originators

People who coined, defined, or first legitimized the concept at scale.

NamePlatform(s)Audience SizeRole in Vibe Coding
Andrej KarpathyX, YouTube~1M+ on XCoined the term “vibe coding” (Feb 2, 2025). Co-founder of OpenAI, former AI lead at Tesla. His single tweet spawned an entire movement, a dictionary entry, and a cottage industry of courses.
Garry TanX, YouTube~500K+ on XCEO of Y Combinator. Reported that 25% of YC Winter 2025 batch had 95%+ AI-generated codebases, massively legitimizing vibe coding as a real startup strategy rather than a novelty.

4. 3. Tier 2: Indie Builder-Celebrities

These influencers gained massive followings by building in public — shipping products rapidly with AI tools and posting revenue screenshots. They embody vibe coding as a lifestyle brand.

NameHandleAudienceProfile
Pieter Levels@levelsio (X)~838K on X The godfather of indie hacking. Built PhotoAI ($106K/mo), InteriorAI ($42K/mo), RemoteOK ($38K/mo), Nomads ($21K/mo). Famous for shipping an AI flight simulator in days. His entire PhotoAI product is a single 40,870-line index.php file. Launched 70+ projects, only 4 made money (<5% hit rate). Appeared on Lex Fridman’s podcast.
Marc Lou@marc_louvion (X)~100K+ on X, YouTube Solopreneur from Bali. ProductHunt Maker of the Year. Shipped 21+ products in 3 years. His biggest revenue driver is ShipFast (~$141K MRR), a NextJS boilerplate — a course/template product, not a vibe-coded app. Practices aggressive build-in-public marketing. Critics note his most profitable product teaches coding, not AI-generated apps.
Greg Isenberg@gregisenberg (X)~400K+ on X CEO of Late Checkout. Former advisor to Reddit and TikTok. Runs “The Startup Ideas Podcast.” Aggressively promotes vibe coding as a SaaS strategy: “vibe code a SaaS, get to $50K/month, sell for 10-15x EBITDA = $3M-$5M.” Also coined “vibe marketing” as the next wave. Publishes practical guides on growing vibe-coded products with creator partnerships.

The Builder Archetype

These influencers share a common playbook: (1) build something fast with AI, (2) post screenshots of the product and revenue on X, (3) attract followers who want the same lifestyle, (4) monetize the audience through courses, boilerplates, and communities. The audience becomes the product.


5. 4. Tier 3: Educator-Amplifiers

These influencers package vibe coding into courses, newsletters, and educational content, reaching massive audiences beyond the developer community.

NamePlatform(s)AudienceProfile
Andrew NgLinkedIn, X, DeepLearning.AILinkedIn Top Voice; millions across platforms Co-founder of Coursera, founder of DeepLearning.AI. Launched the free “Vibe Coding 101 with Replit” course (24,000+ enrolled). Lends massive academic credibility to vibe coding. His endorsement signals to enterprises and professionals that this is a legitimate skill.
Zain KahnLinkedIn, X~505K on X; 2M+ newsletter readers Founder of Superhuman AI, one of the largest AI newsletters (2M+ daily readers). LinkedIn certified Top Voice. Regularly covers vibe coding tools and trends. Has posted about “getting the best vibe coder in the world” to build his app ideas. Monetizes through the newsletter and podcast.
Riley BrownTikTok, X, YouTube~636K on TikTok; 8.9M TikTok likes Co-founder of vibecode.dev. Self-describes as someone who has “never written a line of code.” Creates comprehensive Cursor tutorials (250+ min guides). Defines vibe coding as “using only your voice in Cursor Composer to build apps.” The non-coder teaching non-coders — the purest expression of vibe coding’s democratization promise.
McKay WrigleyX, YouTube~200K+ on X Founder of Takeoff AI. Creates educational content on AI coding tools. Showcases Cursor, Gemini, and voice-driven development workflows. Known for high-quality demo videos that go viral. Bridges the gap between developer-focused content and mainstream AI hype.

6. 5. Tier 4: Tool Evangelists & Platform Voices

People who promote vibe coding through their roles at (or close association with) the major AI coding platforms.

NameAffiliationPlatform(s)Role
Logan KilpatrickGoogle (AI Studio)X, YouTube Leads product at Google AI Studio. Previously at OpenAI, Apple, NASA. Actively promotes Google AI Studio’s “vibe coding” Build feature — prompt-to-production with Gemini. Announced free voice AI agent vibe coding in AI Studio.
Felix (Lovable)LovableLinkedIn, X Posts about Lovable’s tools and integrations. Shares AI content creation tips, walkthroughs, and design trends. Lovable itself hit $100M ARR in 8 months, making it one of the fastest-growing vibe coding platforms.
Michele Catasta & Matt PalmerReplitLinkedIn, Replit community President and Head of DevRel at Replit, respectively. Teach the Andrew Ng “Vibe Coding 101” course. Replit’s Agent v2/v3 and Design Mode have been major catalysts for the movement.
Ray VillalobosLinkedIn LearningLinkedIn LinkedIn Top Voice. Teaches “Getting Started with Vibe Coding” and “Vibe Coding Fundamentals” on LinkedIn Learning. Brings vibe coding to the corporate/professional audience through LinkedIn’s platform.

7. 6. TikTok & YouTube Creators

Vibe coding has exploded on short-form video platforms, often reaching audiences with zero programming background. Key creators include:

  • Riley Brown (@rileybrown.ai on TikTok) — 636K followers, 8.9M likes. Posts viral “build a video game in 2 minutes” demos.
  • @aisavvy (TikTok) — Posts concise explainers (“Everything explained in 89 seconds”) on vibe coding tools and dangers.
  • Marc Lou (YouTube) — Build-in-public vlogs showing his product launches and revenue dashboards. Heavy influence on the “digital nomad coder” aesthetic.
  • McKay Wrigley (YouTube) — Technical demos of Cursor, Gemini, and AI coding workflows. More developer-oriented than other creators.
  • Educative Inc. (@educativeinc on TikTok) — Institutional content explaining vibe coding for software developers transitioning to AI-assisted workflows.

The TikTok vibe coding content tends to focus on spectacle (“I built an app in 2 minutes!”) while YouTube content is longer-form and tutorial-oriented. LinkedIn content skews toward enterprise adoption and career implications.


8. 7. The Counter-Voices & Critics

Not everyone is cheerleading. A growing counter-movement calls out the hype:

SourceCritique
theSeniorDev (“Vibe-Coding Influencers Are Ruining Your Life”) Argues influencers like Pieter Levels and Marc Lou are primarily great at marketing, not engineering. Points out Marc Lou’s most profitable product is a course about coding, not an app. Notes that influencers hide “67 failed apps” and months earning $0 behind curated revenue screenshots.
Simon Willison Distinguishes between vibe coding (casual, accept-everything approach) and serious AI-assisted programming. Argues “not all AI-assisted programming is vibe coding” — professionals should review generated code carefully.
Andrej Karpathy (the inventor himself) By early 2026, declared vibe coding “passé” and notably hand-coded his own new project (tinymorph), admitting AI coding wasn’t sufficient for serious work.
Paweł Brodziński (Substack) “Vibe-Coded Product Success Stories Ain’t What They Look” — examines survivorship bias in the vibe coding narrative.
Daniel Bentes (LinkedIn) “The Reality of Vibe Coding” — provides a professional engineering perspective on the limitations and risks.

9. 8. The Economics: What They Actually Sell

A critical insight: the biggest vibe coding influencers rarely make their money from vibe-coded products. Here’s what they actually monetize:

Revenue StreamExamplesEstimated Revenue
Boilerplates & TemplatesMarc Lou’s ShipFast (NextJS boilerplate)~$141K/month MRR
NewslettersZain Kahn’s Superhuman AI (2M+ subscribers)Estimated $1M+/year (sponsorships)
CoursesAndrew Ng’s DeepLearning.AI courses; Riley Brown’s vibecode.dev; LinkedIn Learning coursesVaries; DeepLearning.AI courses are free (lead gen for platform)
Communities & CohortsMcKay Wrigley’s Takeoff AIUndisclosed
PodcastsGreg Isenberg’s Startup Ideas Podcast; Superhuman AI podcastSponsorship-driven
Actual SaaS ProductsPieter Levels’ PhotoAI, InteriorAI, RemoteOK, Nomads~$207K/month combined (the exception, not the rule)

The pattern: Most vibe coding influencers follow a flywheel: build something fast → post about it → grow audience → sell the audience courses/templates on how to do the same thing. The audience is the real product. Pieter Levels is the notable exception — he makes serious money from actual software products.


10. 9. Platform Landscape Powering the Movement

The tools these influencers promote and build with:

PlatformTypeNotable Metric
CursorAI-powered code editor (IDE)Parent company Anysphere valued at $9.9B (June 2025)
LovableAI app builder (browser-based)$100M ARR in 8 months
ReplitCloud IDE with AI AgentAgent v2, v3, Design Mode throughout 2025
Bolt.newBrowser-based AI app builderKnown as the “speed demon” — idea to shareable URL in minutes
v0 (Vercel)AI UI generationTightly integrated with Next.js ecosystem
Google AI StudioPrompt-to-production with GeminiFree vibe coding experience; voice AI agent builder
Claude Code (Anthropic)CLI-based AI coding agentGrowing adoption for “agentic engineering” workflows

11. 10. Key Takeaways

  1. The term went from tweet to dictionary in 13 months. Karpathy’s Feb 2025 tweet became Collins’ Word of the Year by end of 2025. Few tech terms achieve cultural velocity this fast.
  2. The biggest influencers sell shovels, not gold. Marc Lou sells ShipFast ($141K/mo), not vibe-coded apps. Zain Kahn sells newsletter sponsorships, not software. The recurring pattern: teach people to vibe code > build vibe-coded products yourself.
  3. Pieter Levels is the real outlier. ~$207K/month from actual running software products, not courses or templates. But even he had a <5% hit rate across 70+ projects.
  4. The inventor already moved on. Karpathy declared vibe coding “passé” and hand-coded his next project. The term he coined may have outgrown his definition.
  5. Platform economics are enormous. Cursor ($9.9B valuation), Lovable ($100M ARR in 8 months). The tool makers are capturing far more value than the vibe coders using them.
  6. YC validation was the inflection point. Garry Tan reporting 25% of YC W25 batch had 95%+ AI-generated code shifted vibe coding from “novelty” to “legitimate strategy.”
  7. Survivorship bias is rampant. Every revenue screenshot hides dozens of failed projects and months of $0 income. Critics like theSeniorDev and Paweł Brodziński are gaining traction pushing back on the narrative.
  8. The audience split is clear. TikTok = spectacle and non-coders. X = indie hackers and builders. LinkedIn = professionals and enterprises. YouTube = tutorials and deep dives. Each platform attracts a different segment of the vibe coding audience.