Zoltan Szogyenyi
Chief builder at Bergside, creator of Flowbite and TypeUI
Co-founder and CEO of Bergside, a bootstrapped two-person company in Timisoara that builds and acquires digital products used by millions. Started with outsourcing, pivoted to templates on ThemeForest (first one rejected), then hit a spike on Black Friday selling Tailwind and Figma design systems. That work became Flowbite, an open-source UI component library with 30M+ npm downloads and 9.1k+ GitHub stars, and a peak near 70k a month on one-time sales. AI later cut that revenue by about 90%. In February 2026 he and co-founder Robert Tanislav started TypeUI, which sells design skills for coding models and analytics for the sites those models ship. TypeUI now does about $4,000 to $6,000 a month. Believes in premium pricing, community-driven distribution, and never doing discounts.
Journeys
Zoltan Szogyenyi
Chief builder at Bergside, creator of Flowbite and TypeUI
From a Rejected ThemeForest Template to 70k a Month
How Flowbite grew an open-source component library into a bootstrapped, two-person business
Zoltan Szogyenyi
Chief builder at Bergside, creator of Flowbite and TypeUI
When the Code Got Cheap
How TypeUI became the business after AI cut the price of Flowbite's components
Events
September 17, 2026
Cowork Timisoara - The Office, Timisoara
Indie TM #17: When the Code Got Cheap
What we learned at our seventeenth meetup, where Zoltan came back with TypeUI after AI cut Flowbite's revenue, and the room told him to measure the path and ask on day 30
April 3, 2026
Cowork Timisoara - The Office, Timisoara
Indie TM #6: From ThemeForest Rejects to 70k/Month
What we learned at our sixth meetup at Cowork Timisoara
Advice & Tips
MindsetA rejection is a data point, not a verdict
Bergside's first ThemeForest template was rejected, and the company that grew out of it now does 70k a month. The rejection was information about one submission, not a judgment on the whole idea. Most founders treat an early no as proof the plan is wrong and quit one iteration before the thing that works. Zoltan and his co-founder treated it as feedback, shipped the next version, and kept shipping. Separate the verdict on a single attempt from the verdict on the direction, and keep going.
ProductLet one big spike tell you where the market is
Bergside sold one Tailwind and Figma kit for 3,000 euros in a single Black Friday night, and that one spike redirected the entire company toward Tailwind components. A sudden, sharp sales response is the market telling you where the demand actually lives, in a way no survey or roadmap meeting ever will. When something you ship spikes far beyond the rest, do not treat it as a lucky one-off. Treat it as a map. Drop what is flat and pour your effort into the thing the market just voted for with its wallet.
DistributionOpen source can be your distribution channel
Flowbite's free open-source library has over 30 million npm downloads, and those downloads are the marketing. Instead of buying attention, Bergside gives away a genuinely useful core and lets adoption carry the brand into millions of projects. The developers who already build with the free components are the warm audience for the paid pro components and framework integrations. If your product can have a free core that people install and depend on, the open-source version is not lost revenue, it is the cheapest and most durable distribution you will ever own.
Business & LegalGive away the core and sell the extras once
Flowbite reached 70k a month with no recurring revenue at all: the library is free, and pro components, sections, and framework integrations were one-time purchases. A free core removes the adoption decision, and one-time pricing removes the renewal anxiety that makes buyers hesitate. That run is the proof that you do not need a subscription to build a real business. Pick the smallest valuable thing to charge for, sell it once at a fair price, and let a large free top-of-funnel feed it.
MindsetOwn the layer AI coding tools still leave missing
TypeUI is a bet that as AI generates more of the code, the scarce thing becomes the consistent design layer on top of it. The general pattern for founders in 2026: do not compete with the AI on the work it does well, find the adjacent layer it leaves missing and own that. AI can generate a hundred components; it cannot yet make them feel like one designed product. Look at where the new tools are strong, then build for the gap they create rather than the one they fill.
MindsetWhen the model writes the code, sell the rules it still gets wrong
Flowbite's paid components fell 90 to 95 percent once models could write that code, while the free library kept doing more than 500,000 downloads a week. The installs proved the brand. They did not pay the invoices. When a model makes your core cheap, the scarce layer moves. Zoltan moved it to design rules the model has to follow, and to the analytics that say whether the page works. Look at the line item a model just made free. Price the next thing it still gets wrong.
ProductPut the design system in a markdown file the model can read
TypeUI's design skills are markdown: buttons, navigation, spacing, accessibility, written by a co-founder with fifteen years in UI and UX. Zoltan would rather give a coding model that file than start in Figma. A Figma board is a picture the model has to guess at. A markdown skill is an instruction it can follow. If you sell taste to people who generate code, write the taste down in the format the model already reads. Then charge for the file, the motion, the sound, and the audit, not for another component zip.
ProductA color change is not a test. Watch the path.
TypeUI could keep two Airbnb-style sites feeling like the same product after a color change, and still lose 79.82% of signups by sending people to pricing before login. The palette held. The path did not. Heatmaps, dead clicks, rage clicks, and scroll depth are there so a traffic number has to explain itself. If you ship a redesign, read the step where people stop before you congratulate the visual system. A consistent feeling is not a conversion.
SEOWrite the page you want a model to cite
TypeUI already gets 34% of its visits from Google, 25% direct, and 10.9% from Reddit, on a 0.3% to 0.5% conversion. The room's next content bet was AIO: posts and comparison tables written so a model quotes them when someone asks what to use. A blog that only chases a keyword misses the answer box. Name the alternatives, put the differences in a table, and make the page specific enough that a model can lift it. Search still sends people. The citation is the new distribution.
DistributionShip a plugin where the buyer is already working
Zoltan's guerrilla channel was a Chrome extension and a Figma plugin: a backlink, plus a person who is already trying to design. That is higher intent than a post hoping someone wanders in from Google. If your buyer lives inside one tool all day, put a small useful piece of your product there. The install is the introduction. The subscription can come after they have felt the time you saved.
Business & LegalAsk on day 30, on a scale from 1 to 10
TypeUI is at $4,000 to $6,000 a month, and the room did not want another feature before a score. Ask customers, from 1 to 10, if they would recommend it. Ask again when the first month ends, including the people who cancel. A showcase of shipped customer sites does the public proof. A watermark on the output is a tax. The day-30 answer tells you whether the $30 plan bought back the hours you promised, while the person still remembers the first month.
ProductUse search demand to decide what features to build
Before building a new feature or component, check what people are actually searching for. Use tools like Semrush to find keywords with real demand. When Flowbite saw search volume for "avatar tailwind" and "datepicker tailwind," they built those components and captured the traffic. Let search data guide your roadmap instead of guessing what users want.
DistributionSpend on influencers instead of ads
If your product has any traction, skip paid ads and invest in influencer marketing instead. Pay 1,000 to 2,000 euros per influencer, feature them on your homepage hero section, and prefer American or Australian creators for the English-speaking market. One influencer creates social proof that attracts others. The ROI on a single well-placed creator video outperforms most ad campaigns for developer tools.
Business & LegalNever discount your prices
Resist the urge to compete on price or offer discounts. Low prices scare away serious clients who associate cost with quality. Once you start discounting, customers learn to wait for sales instead of buying at full price. In the digital products space, maintaining premium pricing attracts better customers and builds a more sustainable business. Flowbite never discounts, and it has not hurt growth.
DistributionPost stories on Reddit, not sales pitches
Reddit rewards authenticity and punishes promotion. When Flowbite launched a new datepicker component, they posted the story of building it, not a sales pitch. Share the journey, the technical decisions, the problems you solved. On X/Twitter, sharing MRR numbers gets engagement. On Reddit, the same post gets buried. Each platform has its own language. Learn to speak it.
SEOWrite on Medium and dev.to with canonical links to your site
Publish articles on Medium and dev.to to reach their built-in audiences, but always set the canonical URL to point back to your own blog. This way you get distribution from the platform while search engines credit your domain as the original source. The content should genuinely help the reader, not be a thinly disguised ad. Spammy content gets flagged on both platforms.
ProductSet up analytics before you start scaling traffic
Do not invest in scaling traffic until you have analytics running. Use Microsoft Clarity (free) or Hotjar for heatmaps, rage click detection, and session replays. Watch your bounce rate and average time on site (one minute is a reasonable benchmark to start). If users leave immediately, more traffic just means more people leaving. Fix the experience first, then scale.