No code, five months
Vadim Strizheus is 19. At 17, while still in high school, he landed a job at a marketing startup in his city. He had been hustling since he was a kid — lemonade, then slime in elementary school — and he always wanted to make his own dollar. In March 2026 he quit that job to build software full-time. Before that year he had never shipped an app, and he had no coding experience at all.
What he built is Vugola, an AI video editing agent. Paste in a long video — a podcast, a livestream, a YouTube link — and it chops it into short, captioned, vertical clips ready to post. The customers are clippers, content creators, agencies and brands with long-form content. Plans run $19.99, $49.99 and $99.99 a month, tiered by AI processing credits.
He built it over about four months, initially prompting Lovable, then asking Claude to help him move off Lovable and learn the basics, and today he builds with OpenAI's Codex. The stack behind it is GitHub, Supabase, Vercel, Cloudflare, PostHog, Datafast and AWS for video rendering. He began marketing in March, after noticing podcast, gaming and livestream clips filling his feeds.
X first, then the 45-minute loop
The first money came from X. Strizheus built a personal brand there first, then attached Vugola to whatever AI topic was going viral that week — an OpenClaw plugin, a video tied to Vugola, anything linked to a trending format. His first $2,000 to $3,000 a month came purely from X, and the app sat there for a while, a commodity clipper in a crowded market.
Then he found the second channel. He created a brand-new Instagram account, bought a ManyChat subscription for $15 a month, and spent his scrolling time studying competitors' viral UGC campaigns — their formats, their creators, their sounds. He reproduced the formats in detail: the same sound, the same transitions, the length of the hook, the length of the demo, the length of the call to action, recording his own face over the copied structure and swapping in his own product.
He posted three to six times a day. The account grew to about 10,800 followers by the time he was filmed, with one clip reaching 132,000 views and others at 18,000 and 21,000. The whole routine took about 45 minutes a day — and once a format worked, he only swapped the hook and the sound.
The numbers, and the caveats
In a Starter Story interview published on 11 October 2026, Strizheus opened his dashboard on camera and gave the numbers: about $11,200 in monthly recurring revenue in the prior four weeks, about 400 paying users, and tens of thousands of signups — roughly 10,000 by his count. The trajectory ran in two phases: $2,000 to $3,000 a month from X alone, then the jump past $10,000 from Instagram.
These are founder-reported figures, shown on his Stripe dashboard during the recording but not independently audited. The signup count is vague — “a few 10,000” — and the app was about five months past launch at recording, so retention over a longer period is unknown. The interview does not clarify customer acquisition costs or the compute expense of AI video processing, which is the real margin question for a clipping tool.
What is not in doubt is the shape of the business: a solo 19-year-old, no code, a product built with AI tools, and a growth engine that is almost entirely a daily content habit. The product is the easy part to copy. The 45 minutes a day is the part nobody wants to do.
The routine, in his own words
Strizheus describes the work as a loop, not a campaign. On X he calls the first lesson F2F — founder to founder: everyone on the platform is selling founders something, from AEO to SEO to Reddit services, and the way through is to listen to what they are selling and ride the topics people already care about. On Instagram the loop is mechanical: scroll competitors' viral clips, copy the format, record your own face over it, swap the hook and the sound, post, repeat.
The product itself is built for that loop. One typed sentence is enough to make a clip — the minimum cost per video is five cents — so a creator can test a hook cheaply and often. The beta editor he was building at the time of the interview is meant to close the workflow end to end, from long video to scheduled short, so the clip never leaves the app.
He has told the story more than once since the Starter Story interview: a podcast appearance, a short-form video about the routine that grew the Instagram account to 10,800 followers, and a newsletter case study all repeat the same beats — no code, four months of building, X first, Instagram second, 45 minutes a day. The repetition is the point. The routine did not change when the numbers did.
The playbook
- Ship with AI coding tools — Lovable for four months, then Claude to move off it, then Codex. The models are the team when there is no team.
- Ride trends on X first — Build a personal brand on X first, then attach the product to whatever AI topic is going viral that week. The first $2K–$3K came from this alone.
- Copy proven formats, swap the hook — Same demo, same call to action, three to six times a day. Only the hook and the sound change — the format is already proven by someone else.
- Automate follow-up for $15 a month — A $15-a-month ManyChat turns Instagram comments into leads while you are not looking. Cheap automation beats expensive ads.