Validate ideas by launching fast and charging money
Without VC funding, Levels builds mini-startups alone and launches within about two weeks. He validates ideas not just through sign-ups, but by checking if users actually pay with credit cards, accepting that most ideas will fail.
Supporting evidence
Original excerpt
I have an idea in my mind and I build it like a mini startup, and I launch it very quickly, within two weeks or something, of building it. And I check if there’s demand and if people actually sign up and not just sign up, but if people actually pay money.
Levels launched Nomad List as a public, editable Google spreadsheet rather than building a website first. He crowdsourced city data from followers, and only converted it into a website after the spreadsheet went viral and accumulated hundreds of cities worth of data.
Supporting evidence
Original excerpt
Nomad List, which is like this website I made to figure out the best cities to live and work as digital nomads, it wasn’t a website. It launched as a Google spreadsheet. So, it was a public Google spreadsheet anybody could edit.
Levels attributes his use of PHP, HTML, and CSS to familiarity rather than deliberate choice. When his startups grew rapidly, he never found time to learn newer technologies like Node.js, prioritizing shipping over adopting supposedly better languages.
Supporting evidence
Original excerpt
I think it’s accidental, because that’s the thing I knew. I knew PHP, I knew HTML, CSS, because you make websites, and when my startups started taking off, I didn’t have time to… I remember putting on my to-do list like, “Learn Node.js,” because it’s important to switch, because this obviously is much better language than PHP, and I never learned it.
Photorealistic AI models retain NSFW training bias
Base Stable Diffusion struggles with photorealism and anatomy. Fine-tuned models trained on explicit content produce better results but retain that bias, requiring builders of AI imaging startups to actively prompt out unwanted NSFW outputs.
Supporting evidence
Original excerpt
if you look at the photorealistic models that people use now, there’s still core of porn there, of naked people. So I need to prompt out, and everybody needs to do this with AI startups, with imaging, you need to prompt out the naked stuff.
With over a million monthly generated photos, Levels randomly tested generation parameters like step count and sampler on 10% of users. He measured favorites and downloads to statistically determine which settings produced better images before rolling them out universally.
Supporting evidence
Original excerpt
on like 10% of the users, I would randomly test parameters and then I would see if they would, because you can favor the photo or you can download it, I would measure if they favor it or like the photo. And then I would A/B test
Scaling often forces founders into unwanted management roles
Levels avoids raising funding because he dislikes the stressful life it brings. He notes that many heavily funded founders secretly regret becoming managers instead of doing creative groundwork, feeling trapped in CEO roles they don't enjoy.
Supporting evidence
Original excerpt
I have a lot of founder friends and they tell me secretly, with hundreds of millions of dollars in funding and stuff, and they tell me, “Next time, if I’m going to do it, I’m going to do it like you, because it’s more fun, it’s more indie, it’s more chill, it’s more creative.”
Levels describes Remote OK's revenue model, noting B2B job boards can charge premium prices. He added cosmetic upsells like rainbow-colored posts and background images for up to $1,000 each, reaching $140K monthly revenue at peak before macroeconomic changes reduced income significantly.
Supporting evidence
Original excerpt
You start with 2.99, but at the peak, when the feds started printing money like 2021, I was making 140K a month with Remote OK with just job posts. And I started adding crazy upsells, like rainbow-colored job posts. You can add your background image. It’s just upsells, man. And you charge a thousand dollars for an upsell.
Levels argues that heavy European regulation creates regulatory capture, benefiting large incumbent companies that can afford compliance while blocking newcomers. He contrasts this with America, where he claims he can start an AI startup immediately by just opening his laptop.
Supporting evidence
Original excerpt
regulation is very good for big companies because they can follow it. I can’t follow it, right? If I want to start an AI startup in Europe now, I cannot because there’s an AI regulation that makes it very complicated for me. I probably need to get notaries involved.
nafyi.
