The Twenty Minute VC (20VC): Base44’s Maor Shlomo on Vibe Coding, the Future of SaaS, and AI Market Dynamics

Podcast: The Twenty Minute VC (20VC)
Host: Harry Stebbings
Guest: Maor Shlomo, Co-Founder of Base44
Date: November 24, 2025
Topic: Why “vibe coding” will disrupt SaaS, defensibility in AI app builders, competition with model providers, and the changing structure of the startup/enterprise software market.

Episode Overview

In this provocative and wide-ranging episode, Harry Stebbings sits down with Maor Shlomo, co-founder of Base44, an AI-powered application builder that sold to Wix for $80M and has since skyrocketed to over $100M in revenue. The conversation traverses the rise of “vibe coding” platforms, the death of bloated SaaS, the coming era of customizable, user-owned software, and the brutal market dynamics of building on frontier LLMs.

Maor challenges conventional wisdom on SaaS moats, margin structures, and the supposed inevitability of commoditization, while shining a light on why he’s more worried about Google than competitors like Replit or Lovable. The two dig deep into investing frameworks for the AI era, sustainable company building, and what the next decade of software might look like.

Key Themes & Discussion Points

1. Why Sell Base44 to Wix?

[04:49 – 07:26]

Bootstrap Beginnings: Maor started Base44 as a fun side project, “I wanted to get back to coding... the traction was insane... the business was very profitable.”
Sell vs. Scale: He was at a crossroads: raise capital and go aggressive, stay solo and risk being outcompeted, or “find a parent company that could push it forward.”
Why Wix? Deep alignment in “spirit and target audience.” The deal was structured so the team could remain “lean, small, and startupy” while leveraging Wix’s support, marketing, and operations.
Vision: The ultimate goal was “building something that matters... that actually made an impact on the world” with much higher odds under Wix.

“It was either I raised a lot of money... or I go and join Wix, that basically triples my chances of building something special.”
— Maor Shlomo [06:53]

2. The End of SaaS as We Know It: “Vibe Coding” & Custom Software

[07:45 – 13:54]

Software Becomes Liquid: Maor predicts the demise of one-size-fits-all SaaS:
- “It will be easier to build your own Salesforce-type CRM than buy a license off the shelf.” [07:52]
- The real value: highly tailored, lean software—no feature bloat, user owns code/data, reduced vendor lock-in.
Use Case Advocacy: Even for non-developers (e.g., small business owners), AI-generated apps will replace hard-to-use no-code builders.
- “Bloated SaaS of the pre-2022 era... it’s going to be harder for people to use.” [12:51]
Skepticism from Harry: Harry challenges whether SMBs care about owning their data or software.
Maor’s Rebuttal: What they really want is simplicity and exact fit to their business process, which AI/vibe coding can finally provide.

“I think small businesses specifically would want to build something leaner and more customized... bloated SaaS of the pre-2022 era: it's going to be harder for people to use.”
— Maor Shlomo [12:46]

3. The Moat Debate: Is Vibe Coding Defensible?

[13:54 – 17:08]

Moat Mythbusting: Yes, AI-powered app builders can be cloned quickly—for simple use cases.
- “It’s relatively easy to create a vibe coding tool. It’s very, very hard to create a platform that could help people build products they'll actually use.” [14:22]
Deep Moat Lies in Complexity: For Maor, the defensibility isn’t in spinning up clones, but in
- Handling “complex... real world use cases”
- “Heavy lifting” backend (integrations, user management, compute, etc.)
- Creating a full product-building ecosystem.
Commoditization of Frontend: Websites and landing pages will get commoditized fastest—real traction (and higher margins) lies in organizational/power user apps.

4. Mapping the Market Future: Who Wins & Who Loses?

[17:08 – 23:16]

Market Segmentation: The space is fragmented between pure AI builders (Base44, Lovable, Replit), traditional SaaS incuments (Salesforce, Atlassian), and pro dev tools (Cursor, Cognition).
Incumbents Adapt: Expect SaaS platforms (like Monday.com or Salesforce) to deeply integrate vibe coding and allow “liquid” customization.
Vertical Integration as Moat: Base44’s “built-in backend” (not just frontend automation) is a key differentiator—hard to copy, hard to migrate away from.
Who is Maor Worried About? Not Replit/Lovable/Bolt—but Google:
- If one model provider “wins by a wide margin,” they’ll move up the stack and conquer the vibe coding market.

“If at one point one of the model providers will just win the market by a wide margin, then the next logical thing is to conquer the vibe coding market.”
— Maor Shlomo [23:10]

5. AI Market Dynamics: LLMs, Margins, and Switching Costs

[23:52 – 31:38]

Fast, Frictionless Switching: Unlike the cloud market, vibe coding tools can switch LLM providers “with a line of code in seconds” [25:59].
Implications: Massive revenue swings for Anthropic, OpenAI, Google, etc., if a new LLM leapfrogs the rest.
Margin Structure: AI model cost is a big % of COGS, but improving:
- “Margins are the least thing that I’m worried about” since model prices are dropping, smaller/faster models can handle simple prompts, and routing is getting smarter.
- Strategic bet: Bet that “model prices will go down toward zero.”
Velocity & Product Iteration: Because copying is so easy, Base44 focuses on building things “very hard to replicate” (homegrown backend) and shipping smaller features fast.

6. Investment Frameworks for the AI Era

[36:35 – 44:59]

Revenue and Moats: Investors are dazzled by huge early revenue—but Maor cautions to look for vertical integration, infrastructure complexity, and the risk of getting “eaten by model providers.”
Defending Against Commoditization: Companies that are only “prompting LLMs” with no deep infrastructure are especially at risk.
Zero to Ten: “0 to 10 million in a year is an insane goal for any SaaS business,” but is increasingly the yardstick for excitement.
Angels in Dull Places: Maor advocates investing more in “not sexy” industries that can be vertically integrated and deeply improved with AI (healthcare, law, finance, restaurants).

7. Company Building & Emotional Resilience

[66:52 – 68:18]

Scarcity to Surplus: Maor recalls the terror of scaling as a solo founder and the emotional impact of early failures and outages: “I took it very, very hard... I wish I could have told myself: this is just going to be a bump in the road.”
Focus: “Don’t get distracted by what competitors are doing... focus on your business, focus on building a great product.”

Notable Quotes & Soundbites

On SaaS Obsolescence:

“I think the current model of one size fits all doesn’t make a lot of sense. It's not that everyone will move there, it’s not that enterprises, it’s going to take a lot of time, but it makes so much sense for the buyer.”
— Maor Shlomo [10:13]
On Moats in AI Builders:

“The mode in the category is being able to get there. And that’s like layers of integrations and heavy lifting and building compute... you’re almost building like a small cloud.”
— Maor Shlomo [15:18]
On Model Switching:

“It’s insane how easy it is... in a split second you change one string in the code and you move millions in spend to a different provider.”
— Maor Shlomo [24:20]
On Defensibility and Copying:

“For us, every feature that we put out... we know it’s going to take either a few weeks or a few months for a competitor to copy.”
— Maor Shlomo [31:40]
On Startups Getting ‘Eaten’:

“If your company is just about how you're prompting an LLM... it's going to be hard to have a moat.”
— Maor Shlomo [44:11]

Quickfire Round Highlights

% of Code Written by AI in 2 Years:
“95%—oh, 100% for sure. You’ll be prompting agents.” [58:17]
Advice to Non-Technical Builders:
“The biggest hack is to build something for your own problems... iterate very fast; it’s OK to throw out thousands of lines of code. It’s not about planning, it’s about iteration.” [59:04]
OpenAI at $500B or Anthropic at $360B?
“Anthropic every day. Moving faster. Better models for growth.” [62:59]
Biggest Misconception at the Frontier Labs:
“Maybe the moat that they have... it takes one better model or lower price to drastically reduce your revenue.” [64:49]
Partner Advice:
“It shouldn’t be ‘this person completes me,’ it should be someone... identical to you... It’s very hard to date a founder, so having a partner with the same mindset is crucial.” [65:51]
Advice to His Past Self:
“Don't take the bumps in the road [too hard]... consumer market feedback is much more emotional than B2B. Handle it better emotionally.” [67:06]

Important Timestamps

[04:49] — Maor explains why he sold Base44 to Wix
[07:45] — Discussion on not regretting the sale; vision for “vibe coding” as the future
[09:00] — Liquid, user-owned software vs. one-size-fits-all SaaS
[13:54] — Are vibe coding tools defensible or easily cloned?
[17:08] — Mapping the market: upmarket/downmarket threats
[23:10] — Maor’s biggest worry is model providers, especially Google
[24:43] — Switching LLMs is frictionless (“a line of code”)
[29:12] — Margin structures at AI app builders
[31:38] — How rapid copying changes product strategy
[42:31] — What revenue growth is now exciting for SaaS investors?
[44:11] — Criteria for defensibility and moats in “AI wrapper” startups
[47:53] — How Base44 aims to beat pure dev tooling like Cursor
[52:06] — Will there be an AI bubble or speed bump? (Maor: “No, we’re only scratching the surface.”)
[58:17] — % of code written by AI in two years (near 100%)
[59:04] — Advice for non-technical builders entering the space
[62:59] — Long Anthropic, not OpenAI? (“Anthropic every day”)
[67:06] — Maor’s advice to his past self: Emotional resilience

Final Reflections

Maor Shlomo’s perspective is a bracing counterpoint to common narratives about low-margin, low-moat AI startups. He’s bullish on deep vertical moats, hybrid go-to-market (with large partners), and the radical expansion (not compression) of the software market as AI collapses complexity and activates new users. Investors and founders alike should take heed: the winners in vibe coding won’t just automate frontends, but deliver infrastructure power, speed, and user empathy usually reserved for traditional enterprise builds.

For more 20VC content: www.20vc.com

The Twenty Minute VC (20VC): Base44’s Maor Shlomo on Vibe Coding, the Future of SaaS, and AI Market Dynamics

Episode Overview

Key Themes & Discussion Points

1. Why Sell Base44 to Wix?

[04:49 – 07:26]

Bootstrap Beginnings: Maor started Base44 as a fun side project, “I wanted to get back to coding... the traction was insane... the business was very profitable.”
Sell vs. Scale: He was at a crossroads: raise capital and go aggressive, stay solo and risk being outcompeted, or “find a parent company that could push it forward.”
Why Wix? Deep alignment in “spirit and target audience.” The deal was structured so the team could remain “lean, small, and startupy” while leveraging Wix’s support, marketing, and operations.
Vision: The ultimate goal was “building something that matters... that actually made an impact on the world” with much higher odds under Wix.

“It was either I raised a lot of money... or I go and join Wix, that basically triples my chances of building something special.”
— Maor Shlomo [06:53]

2. The End of SaaS as We Know It: “Vibe Coding” & Custom Software

[07:45 – 13:54]

Software Becomes Liquid: Maor predicts the demise of one-size-fits-all SaaS:
- “It will be easier to build your own Salesforce-type CRM than buy a license off the shelf.” [07:52]
- The real value: highly tailored, lean software—no feature bloat, user owns code/data, reduced vendor lock-in.
Use Case Advocacy: Even for non-developers (e.g., small business owners), AI-generated apps will replace hard-to-use no-code builders.
- “Bloated SaaS of the pre-2022 era... it’s going to be harder for people to use.” [12:51]
Skepticism from Harry: Harry challenges whether SMBs care about owning their data or software.
Maor’s Rebuttal: What they really want is simplicity and exact fit to their business process, which AI/vibe coding can finally provide.

“I think small businesses specifically would want to build something leaner and more customized... bloated SaaS of the pre-2022 era: it's going to be harder for people to use.”
— Maor Shlomo [12:46]

3. The Moat Debate: Is Vibe Coding Defensible?

[13:54 – 17:08]

Moat Mythbusting: Yes, AI-powered app builders can be cloned quickly—for simple use cases.
- “It’s relatively easy to create a vibe coding tool. It’s very, very hard to create a platform that could help people build products they'll actually use.” [14:22]
Deep Moat Lies in Complexity: For Maor, the defensibility isn’t in spinning up clones, but in
- Handling “complex... real world use cases”
- “Heavy lifting” backend (integrations, user management, compute, etc.)
- Creating a full product-building ecosystem.
Commoditization of Frontend: Websites and landing pages will get commoditized fastest—real traction (and higher margins) lies in organizational/power user apps.

4. Mapping the Market Future: Who Wins & Who Loses?

[17:08 – 23:16]

Market Segmentation: The space is fragmented between pure AI builders (Base44, Lovable, Replit), traditional SaaS incuments (Salesforce, Atlassian), and pro dev tools (Cursor, Cognition).
Incumbents Adapt: Expect SaaS platforms (like Monday.com or Salesforce) to deeply integrate vibe coding and allow “liquid” customization.
Vertical Integration as Moat: Base44’s “built-in backend” (not just frontend automation) is a key differentiator—hard to copy, hard to migrate away from.
Who is Maor Worried About? Not Replit/Lovable/Bolt—but Google:
- If one model provider “wins by a wide margin,” they’ll move up the stack and conquer the vibe coding market.

“If at one point one of the model providers will just win the market by a wide margin, then the next logical thing is to conquer the vibe coding market.”
— Maor Shlomo [23:10]

5. AI Market Dynamics: LLMs, Margins, and Switching Costs

[23:52 – 31:38]

Fast, Frictionless Switching: Unlike the cloud market, vibe coding tools can switch LLM providers “with a line of code in seconds” [25:59].
Implications: Massive revenue swings for Anthropic, OpenAI, Google, etc., if a new LLM leapfrogs the rest.
Margin Structure: AI model cost is a big % of COGS, but improving:
- “Margins are the least thing that I’m worried about” since model prices are dropping, smaller/faster models can handle simple prompts, and routing is getting smarter.
- Strategic bet: Bet that “model prices will go down toward zero.”
Velocity & Product Iteration: Because copying is so easy, Base44 focuses on building things “very hard to replicate” (homegrown backend) and shipping smaller features fast.

6. Investment Frameworks for the AI Era

[36:35 – 44:59]

Revenue and Moats: Investors are dazzled by huge early revenue—but Maor cautions to look for vertical integration, infrastructure complexity, and the risk of getting “eaten by model providers.”
Defending Against Commoditization: Companies that are only “prompting LLMs” with no deep infrastructure are especially at risk.
Zero to Ten: “0 to 10 million in a year is an insane goal for any SaaS business,” but is increasingly the yardstick for excitement.
Angels in Dull Places: Maor advocates investing more in “not sexy” industries that can be vertically integrated and deeply improved with AI (healthcare, law, finance, restaurants).

7. Company Building & Emotional Resilience

[66:52 – 68:18]

Scarcity to Surplus: Maor recalls the terror of scaling as a solo founder and the emotional impact of early failures and outages: “I took it very, very hard... I wish I could have told myself: this is just going to be a bump in the road.”
Focus: “Don’t get distracted by what competitors are doing... focus on your business, focus on building a great product.”

Notable Quotes & Soundbites

On SaaS Obsolescence:

“I think the current model of one size fits all doesn’t make a lot of sense. It's not that everyone will move there, it’s not that enterprises, it’s going to take a lot of time, but it makes so much sense for the buyer.”
— Maor Shlomo [10:13]
On Moats in AI Builders:

“The mode in the category is being able to get there. And that’s like layers of integrations and heavy lifting and building compute... you’re almost building like a small cloud.”
— Maor Shlomo [15:18]
On Model Switching:

“It’s insane how easy it is... in a split second you change one string in the code and you move millions in spend to a different provider.”
— Maor Shlomo [24:20]
On Defensibility and Copying:

“For us, every feature that we put out... we know it’s going to take either a few weeks or a few months for a competitor to copy.”
— Maor Shlomo [31:40]
On Startups Getting ‘Eaten’:

“If your company is just about how you're prompting an LLM... it's going to be hard to have a moat.”
— Maor Shlomo [44:11]

Quickfire Round Highlights

% of Code Written by AI in 2 Years:
“95%—oh, 100% for sure. You’ll be prompting agents.” [58:17]
Advice to Non-Technical Builders:
“The biggest hack is to build something for your own problems... iterate very fast; it’s OK to throw out thousands of lines of code. It’s not about planning, it’s about iteration.” [59:04]
OpenAI at $500B or Anthropic at $360B?
“Anthropic every day. Moving faster. Better models for growth.” [62:59]
Biggest Misconception at the Frontier Labs:
“Maybe the moat that they have... it takes one better model or lower price to drastically reduce your revenue.” [64:49]
Partner Advice:
“It shouldn’t be ‘this person completes me,’ it should be someone... identical to you... It’s very hard to date a founder, so having a partner with the same mindset is crucial.” [65:51]
Advice to His Past Self:
“Don't take the bumps in the road [too hard]... consumer market feedback is much more emotional than B2B. Handle it better emotionally.” [67:06]

Important Timestamps

[04:49] — Maor explains why he sold Base44 to Wix
[07:45] — Discussion on not regretting the sale; vision for “vibe coding” as the future
[09:00] — Liquid, user-owned software vs. one-size-fits-all SaaS
[13:54] — Are vibe coding tools defensible or easily cloned?
[17:08] — Mapping the market: upmarket/downmarket threats
[23:10] — Maor’s biggest worry is model providers, especially Google
[24:43] — Switching LLMs is frictionless (“a line of code”)
[29:12] — Margin structures at AI app builders
[31:38] — How rapid copying changes product strategy
[42:31] — What revenue growth is now exciting for SaaS investors?
[44:11] — Criteria for defensibility and moats in “AI wrapper” startups
[47:53] — How Base44 aims to beat pure dev tooling like Cursor
[52:06] — Will there be an AI bubble or speed bump? (Maor: “No, we’re only scratching the surface.”)
[58:17] — % of code written by AI in two years (near 100%)
[59:04] — Advice for non-technical builders entering the space
[62:59] — Long Anthropic, not OpenAI? (“Anthropic every day”)
[67:06] — Maor’s advice to his past self: Emotional resilience

Final Reflections

For more 20VC content: www.20vc.com

wavePod

20VC: Base44's Maor Shlomo on How Vibe Coding Will Kill SaaS and Salesforce | Why it is BS that Vibe Coding Platforms Do Not Have Defensibility and Bad Margins | Why He Worries About Google, Not Replit and Lovable | Why Long Anthropic, Not OpenAI?

Powered by Wave AI

Summary

The Twenty Minute VC (20VC): Base44’s Maor Shlomo on Vibe Coding, the Future of SaaS, and AI Market Dynamics

Episode Overview

Key Themes & Discussion Points

1. Why Sell Base44 to Wix?

2. The End of SaaS as We Know It: “Vibe Coding” & Custom Software

3. The Moat Debate: Is Vibe Coding Defensible?

4. Mapping the Market Future: Who Wins & Who Loses?

5. AI Market Dynamics: LLMs, Margins, and Switching Costs

6. Investment Frameworks for the AI Era

7. Company Building & Emotional Resilience

Notable Quotes & Soundbites

Quickfire Round Highlights

Important Timestamps

Final Reflections

Summary

The Twenty Minute VC (20VC): Base44’s Maor Shlomo on Vibe Coding, the Future of SaaS, and AI Market Dynamics

Episode Overview

Key Themes & Discussion Points

1. Why Sell Base44 to Wix?

2. The End of SaaS as We Know It: “Vibe Coding” & Custom Software

3. The Moat Debate: Is Vibe Coding Defensible?

4. Mapping the Market Future: Who Wins & Who Loses?

5. AI Market Dynamics: LLMs, Margins, and Switching Costs

6. Investment Frameworks for the AI Era

7. Company Building & Emotional Resilience

Notable Quotes & Soundbites

Quickfire Round Highlights

Important Timestamps

Final Reflections