What Will Be the Next "Cursor for ____"?
The meteoric rise of Cursor and what we can learn from it to build the next big thing
At the end of 2024, Cursor took the crown for being the fastest growing SaaS product ever, reaching $100M ARR within 12 months. Now, people can’t stop talking about Cursor; it’s dominating my X and LinkedIn feeds and it’s easily become the highest ROI subscription I personally pay for.
Great startups inspire everyone. From founders to VCs to even those outside the industry, they’re abuzz talking about the crazy new startup changing the world and, more importantly, how they can get in on the action.
It reminds me of my time at Uber during its meteoric rise which inspired so many copycats and offshoots. For a while, I felt like every other founder I came across was pitching “we’re like Uber, but for food” or “we’re like Uber, but for laundry.” And thus the “Uber for X” meme was born and proliferated in the founder/VC world. Which made perfect sense because the novel concept of Uber was so easy to understand and relatable to other verticals.
Now, I think the question is: Who will be the next “Cursor for X”? Like Uber, Cursor’s concept is brilliant and disruptive but not limited to one vertical, especially since AI is the key unlock and it could be applied so broadly. If you Google around, you can already see the meme permeating YC’s new batch and even earlier ones pivoting/rebranding. So, it’s already happening, but it’s still early. Let’s go deeper into it!
Uber’s Rise and Influence
Uber hit 1 billion trips 5.5 years in (Christmas eve, 2024) then hit 2 billion just 6 months later (and 10 billion 2 years after that). What made Uber unique at the time was:
On-demand
Mobile
Physical (location-based)
Marketplace
Regulatory
The first four defined how the product worked and alone they were enough to make it fairly unique, but I think the last one was really what led to massive disruption. Despite how messy it was, Uber’s battles with almost every single city they entered emboldened other founders to tackle new markets never before considered (and investors likewise updated their TAM calculations assuming they’d found the next Travis Kalanick).
Ultimately, many startups emerged out of this wave and a good amount saw great success. Excluding successful direct competitors in ride hailing (Didi, Grab, Ola), the winners were all delivery/food:
DoorDash - Founded: 2013, Valuation: ~$74B
Instacart - Founded: 2012, Valuation: ~$11B
Deliveroo - Founded: 2013, Valuation: ~$2.6B
Rover - Founded: 2011, Exit: $2.3B (acquired by Blackstone in 2024)
Not yet exited, valuation unclear:
Gopuff - Founded: 2013, Last Valuation: ~$15B (Series H in 2021)
Getir - Founded: 2015, Last Valuation: ~$12B (Series E in 2022)
Deconstructing Cursor's Success
First, a timeline to put Cursor’s success into context:
2022 January - Founded by MIT students Aman Sanger, Arvid Lunnemark, Michael Truell, and Sualeh Asif
2023 April - Public Launch
2023 September - Seed Round: $11M; $1M ARR, tens of thousands of users
2024 August - Series A: $60M ($400M valuation)
2024 December - Reached $100M ARR
2025 January - Series B: $105M (~$2.5B valuation)
Current: $200M+ ARR, 360k users (speculated)
That kind of trajectory doesn't happen by accident. So, what drove it? Breaking it down like Uber, Cursor is:
Personal SaaS (one person can get value)
Built for a specialized customer (coders)
Up against existing major incumbent (Github Copilot)
Freemium model
Uses AI
… ?
On the surface, it doesn’t actually sound all that groundbreaking. Sounds like just another cute SaaS tool without first-mover advantage, right? But in a world where Cursor doesn’t exist, I think most investors would pass (indeed many did if you ask around) with the rationale:
Can’t beat the 800lb gorilla (Github/Microsoft)
Small TAM; too specialized
GPT wrapper, just a feature away for OpenAI/Google
Let’s dig deeper into the key factors to their success, which I think are nuanced:
Reduced friction - Engineers commonly relied (past tense!) on Google to get their job done. There’s no ill-intent, it was just a fact of life given how expansive and rapidly growing the software ecosystem is. Even the best developers do this. The impact? StackOverflow’s traffic dropped dramatically. Clearly, it was a problem worth solving.
Familiar yet proprietary - Reducing adoption friction, Cursor wisely chose to fork VS Code which already had millions of users (some estimate 20M when Cursor launched). So, with a freemium model, a virtually frictionless switch, and the promised upside of AI, there was nothing stopping engineers from at least trying it out.
Immersive experience - Engineers commonly get into a flow state where they are intensely focused and building for hours on end. Any disruption could throw them off, so not needing to switch out of their IDE and into a browser (where they might end up hunting down issues for hours) greatly increases their likelihood of staying in a deep state of flow.
Context (iterative vs transactional) - While you could go to ChatGPT and get some coding help, you typically went back to your IDE once you got your answer. In that form (at the time), it was really just better information retrieval with zero context beyond the prompt; a replacement for Googling. However, engineers can spend years in a single codebase, continually improving, refactoring, and building upon it. So being embedded in the IDE itself meant the AI always has context, which means the world in terms of its output quality.
Product led growth - Putting it all together, Cursor pulled off PLG perfectly. Like many engineers, I learned about Cursor from friends and co-workers. They essentially spent zero on marketing and relied solely on word-of-mouth, entirely based on the user experience and the value it delivered. Ironically, you could also say their PLG is flawed because the IDE contains zero annoying nudges to invite or share with others; users really have to go out of their way to spread the word.
High personal ROI - Engineering has become one of the most lucrative careers today, with the best engineers achieving higher total comp than even doctors. So, would an engineer spend $20/mo to increase their chances of becoming a Staff engineer at MAMAA pulling down $500k - $1M LIQUID TC per year? And now that Cursor can effectively take a lot of menial work off the table, what else can they do with their time? Money-making side projects? Yes and yes.
New generation of “coders” - Lowering friction for professionals also benefited non-pros and led to a surge of new builders. Just like Uber expanded the transportation market, Cursor (and AI overall) made the market for builders swell, attracting droves of new startups and innovations.
Timing - You can’t ignore Cursor’s impeccable timing, whether intentional or just lucky. Not too early when LLMs weren’t quite there, but just right when AI for coding had really gotten past the 0 to 1 phase thanks to Github Copilot paving the way. This was critical because many companies were cautious to allow Copilot fearing code might be shared or used for training. Launching after them actually meant Cursor saw even less friction to adoption. Additionally, public sentiment had already began shifting from “be careful about sharing stuff with AI” to essentially “you will die if your employees don’t use AI like NOW!”
Does Cursor Have A Moat?
Clearly, Cursor nailed it as a product. If the organic, word-of-mouth is a testament, then users absolutely love it. But is simply—I use this lightly, I know it’s hard—being an amazing product a defensible moat? If an identical product landed in a user’s lap, and it were say cheaper or meaningfully better/faster, would they just switch? It’s not a slam dunk, but it’s hard to objectively say that there are high switching costs.
Of the 7 powers of business strategy, you could say Cursor’s counter-positioning was key to how they got here and that they have a strong brand thanks to all the public advocacy. But today it’s hard to say that they have a truly, strategically defensible moat without having something new up their sleeves. Copilot has already reached parity at half the price and Windsurf is gaining traction with a potentially sizable war chest from OpenAI. On the other hand, Gemini has seemingly exceeded ChatGPT on objective measures yet ChatGPT still dwarfs them by ~16X, so maybe brand does matter a lot?
Sometimes, obsessing about a moat can be debilitating, so I wouldn’t push on this too much. It’s just healthy to keep in mind.
Building the Next "Cursor for X"
So, my takeaways on what Cursor has shown matters:
Specialized user that relies on specialized knowledge
Where AI is a huge help, but can’t easily replace them (yet)
Highly tailored product built around their flow
Always-on context and iterative work (vs transactional)
Personal value creation; enablement
Frictionless adoption, single-player mode
Distribution hack: VS Code
That last one is probably the most interesting, because if you’re like me you believe that:
AI is now cheap
software is now easy
BUT distribution is still very, very hard
Bringing this all together, let’s explore where the next “Cursor of X” can be.
Popular Open Source Projects
Starting with the hardest problem and working backwards, we see some interesting opportunities to improve an existing product by incorporating AI effectively. Here, I focused on open source projects that were designed for individual user (single-player mode).
I’ll say up front that while the users are indeed specialists, I’m not so sure they face nearly the same unique knowledge challenges that developers do. Can they benefit from AI? Absolutely. And having a distribution hack from the get go gives you you a huge edge so they’re definitely worth looking into.
Cursor for Office
TAM: $30B-$60B
I was surprised to learn that LibreOffice apparently has 200M users worldwide. This is incredibly interesting!
Although the office-flavored AI apps (writing, spreadsheets, presos) space is pretty crowded, the Cursor model of being tightly integrated into a worker’s existing workflow can be a game changer. Even if users did try out those other apps, their transactional nature might have brought them back home to LibreOffice to iterate.
On top of the enterprise nature, the scale tells me there’s a significant portion of the install driven by top-down mandate, so there may be friction in getting folks install your LibreCursor. The key question is how many of these users are individuals with autonomy to choose their own software. Even 5% is very meaningful.
Cursor for Design
TAM: $10B - $25B
GIMP is the most popular open source design app with Krita, Inkscape, and Blender also as viable options. From what I can tell, GIMP may have something like 200M users (!) and the others could be around the 250k-1M mark.
The unique challenge in design is that you’d be up against Figma who has a stellar team and already incorporates AI into their product. However, it’s not a zero sum game and it doesn’t take away from GIMP’s 200M users. There’s something very real to be had here.
Cursor for Audio
TAM: $5B-$10B
Another surprising one: Audacity, supposedly has “hundreds of millions” of downloads. Given audio software can be pricey, this seems promising. While AI in music has been a bit tricky with licensing, AI in audio editing could be incredibly interesting given you’re dealing with professionals.
I spent a little time playing with audio editing, so I can imagine how far some AI could go, not to mention open things up to make way for more creators. Feels like a very meaningful opportunity.
Other: Video, CAD, GIS
TAM: $5B - $25B
Video is growing quickly right now and I think it could be interesting to look at Shotcut, OpenShot, and Kdenlive each with meaningful adoption. While we have OpenAI and Google innovating on video models, it’s nothing like a pro’s workflow which could really benefit from AI. There’s a chance we’re a bit early for video, but maybe it’s actually perfect timing.
Others to consider:
Huge Markets with Specialists
Looking at massive markets where highly specialized users require extensive knowledge—and thus AI can be a huge unlock—we see a harder challenge where the software is deployed professionally rather than by individuals, and the environments can be highly regulated. Hoards of startups are tackling these markets with AI, but the road to adoption much longer without a distribution hack and more complicated with the regulation. There’s still room for a Cursor for X, but it’s absolutely not for the faint of heart.
Cursor for Healthcare
TAM: $ Trillions
AI + healthcare is a hugely popular thesis, so I’ll just get this out the way. However, it’s an incredibly hard nut to crack. Again, the software and AI are not the hard part, but assuming you can convince all the parties involved to install and use your tech, then you could have massive impact on the world.
My instinct is that tackling healthcare would necessarily have to start very narrow and go deep before ever entertaining a broader rollout. I might target roles oriented around diagnosis like radiology or pathology or perhaps data-oriented like coders and billers. Some startups have targeted more administrative roles, but that may fall too far outside the Cursor sweet spot of specialized users who leverage knowledge.
Cursor for Home Services
TAM: $700B - $900B
A company I’m advising works in this space and it’s been exciting brainstorming with them on how AI could disrupt here. Like healthcare and education, there’s an enterprise SaaS ecosystem to navigate, but they’re not too complicated and the PE owners are sophisticated enough to understand how tech and AI can disrupt. On top of that, home service providers like plumbers, roofers, mechanics, etc are a long way from being replaced by AI/robots so they’re not as fearful.
What excites me is that whatever’s built to help the pros would eventually trickle down to the DIYers who love working on their home. Things like replacing your HVAC filters, changing your car’s oil, or navigating the myriad building codes to modify your home could be made effortless for anyone with an AI companion. And the ROI element applies here too; home projects become investments that yield returns in home value, a strong incentive for any homeowner: $20/mo to increase your home’s value by 10%.
Cursor for Education
TAM: $150B - $350B
I’m most excited about how AI can help fix our broken education system. While still regulated and enterprise-ey, it’s not nearly as gnarly as healthcare. The best part is there’s a very popular open source platform call Moodle that has seen substantial adoption: 150k registered sites, 237+ countries, 444M registered users, 51M courses hosted (data sourced from Gemini).
Building for teachers might look CRM-like where you amplify the quality and depth with which they go wide to support as many students as possible. All the menial things they do like grading, reporting, and administrator can be handed off to AI, while their unique abilities to build curriculums and engage with students and parents would be 10X’d by AI.
I would love to talk to startups working on this!
Conclusion
Cursor's story is compelling, especially its rapid scaling driven by a developer-first PLG approach combined with an AI-native product vision. It highlights how a focused strategy, leveraging deep AI integration and potentially counter-positioning, can create significant traction even against giants.
The "Cursor for X" framework is a useful lens for brainstorming where similar deep AI integration could revolutionize professional workflows. But, as we learned from the "Uber for X" era, the real winners won't be those who just copy the label or the initial go-to-market motion. They'll be the founders who deeply understand the specific problems, workflows, risks, and nuances of the industry they're targeting. They'll figure out how to build not just a clever AI tool, but a trustworthy, indispensable partner for professionals, and maybe become the next fastest-growing SaaS company by doing so. The opportunity is huge, but it demands thoughtful, industry-specific innovation.
(Please let me know if you’re working on any of these!)