Y CombinatorVertical AI Agents Could Be 10X Bigger Than SaaS
At a glance
WHAT IT’S REALLY ABOUT
Vertical AI Agents Poised To Eclipse SaaS And Reshape Workforces
- The hosts argue that vertical AI agents—AI systems tailored to specific business functions—could produce at least 300 new billion‑dollar companies, potentially becoming 10x larger than today’s SaaS winners because they replace both software and headcount.
- They draw historical parallels between the rise of SaaS (enabled by browser tech like XMLHttpRequest) and the current LLM wave, emphasizing that most value in the last era accrued to B2B vertical tools rather than obvious consumer apps.
- Vertical AI agents are already gaining traction in domains like QA testing, recruiting, customer support, developer relations, call centers, surveys, and medical/dental billing by fully automating previously human‑heavy workflows.
- The conversation explores why incumbents rarely dominate vertical software, how AI may change company scaling and org design, and how founders should identify promising verticals by targeting boring, repetitive administrative work they understand deeply.
IDEAS WORTH REMEMBERING
5 ideasVertical AI agents can subsume both software and the human teams operating it.
Where SaaS replaced boxed software but still required large ops teams, AI agents combine the app plus the work of the people using it, allowing companies to eliminate entire functions like QA, recruiting screens, or call-center agents.
The SaaS playbook suggests hundreds of vertical AI unicorns are plausible.
Over 300 SaaS unicorns emerged as software moved to the browser; the hosts argue each major SaaS category (e.g., payroll, surveys, QA) can now have an AI‑agent equivalent that is larger because it also eats payroll spend.
Incumbents are poorly positioned to win most vertical AI categories.
Deep domain knowledge, focus, and willingness to overhaul existing products are required for each niche; big tech firms rarely invest in the obscure details and organizational risk needed to dominate many narrow domains.
Founders should chase “boring, repetitive admin work” they know firsthand.
The most promising verticals often emerge from observing mundane, high-volume processes—like medical billing, government RFP bidding, or survey analysis—especially where someone’s full-time job is structured, language-heavy grunt work.
AI will change how startups scale and who they hire post–product-market fit.
Instead of building large go-to-market and ops teams, founders are advised to hire more strong engineers who understand LLMs and can systematically automate growth bottlenecks and internal functions.
WORDS WORTH SAVING
5 quotesLiterally every company that is a SaaS unicorn, you could imagine there is a vertical AI unicorn equivalent in some new universe.
— Jared
The vertical AI equivalent is just gonna be the software plus the people in one product.
— Jared
If you can find a boring, repetitive admin task, there is likely going to be a billion‑dollar AI agent startup if you keep digging deep enough into it.
— Harj
These vertical AI agent companies can be ten times as big as the SaaS companies that they are disrupting.
— Diana
The rocks can talk and can have conversations, but the more interesting thing is that the rocks can read.
— Garry (paraphrasing Parker Conrad)
High quality AI-generated summary created from speaker-labeled transcript.
Get more out of YouTube videos.
High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.
Add to Chrome