Fundraising in the age of AI: why investors now want something else
No Bullsh!t, April 29, 2026, 73 min, 279 views. This episode's guest, Andrey Rezinkin, an angel investor and author of the venture capital channel Money for Startups, shares his path: from mathematician-economist and finance automation contractor at large companies, through his own water-purification business and a family office, to professional angel investing since 2022.
| Published | |
|---|---|
| Length | 73 min |
| Views on YouTube | 279 |
| Language of the episode | Russian |
| Where to watch and listen | YouTube, Spotify, Telegram |
Date, length and view count according to data from the episode page on YouTube, as of the page update date shown in the footer.
In short
He candidly breaks down several of his failures: the American startup AlphaFlow, which didn't survive rising interest rates; the education project ReFocus in Southeast Asia, where the credit-based sales model collapsed; and his first attempt at running a restaurant. The main theme of the conversation is how the venture market has changed in the age of artificial intelligence: the classic pre-seed round for simple tech projects has almost disappeared, because vibe-coding lets founders build an MVP and land first customers fast, while for deep-tech projects pre-seed rounds have survived. The conversation also covers the number of agents used and the volume of tokens burned as a new signal for investors, as well as stablecoins and cross-border payments as a theme underrated amid the AI hype, around which Andrey is raising his own new fund. The episode closes with advice for early-stage investors and founders.
Key ideas
- The classic pre-seed round for simple tech projects has nearly disappeared: vibe-coding lets founders quickly build an MVP and land first paying customers, so investors now expect a seed-stage company with a revenue growth trajectory; for deep-tech projects (a portfolio example, the company Sirin DB), idea-stage pre-seed rounds still happen.
- Investors who specialize in AI use the number of agents a team runs and the volume of tokens it burns as a quantitative signal for initial screening, before moving on to qualitative evaluation.
- Traction is the king: the combination of traction and a team capable of sustaining it and deliberately planning the next round settles the fundraising question more than famous co-investor or accelerator names.
- The biggest mistake a beginner angel makes, from Andrey's own experience, is investing under the influence of high-status co-investors (the AlphaFlow case) rather than one's own analysis; his takeaway: always do your own due diligence and ignore the badges.
- Stablecoins and cross-border B2B payments are a theme underrated amid the AI hype: SWIFT remains expensive and slow (2-5 days, double conversion through the dollar), while blockchain payments take seconds to minutes; traditional fintech is turning this way under pressure from its own clients' demand, not out of conviction.
- Stress-testing your financial model with domain experts (experienced restaurateurs, financial-modeling specialists) before launch saves money and time; the lesson applies to any business, not just venture-backed startups.
- When looking at failed deals, it's important to examine not only the founders' mistakes but also the deal's origin: if an unfamiliar team reaches out cold on its own, that alone is a signal to be cautious, not a reason for interest.
The full transcript
The episode is recorded in Russian, and the full transcript of the conversation is on the Russian page of this episode. English and Spanish subtitles for the strongest episodes are rolling out on YouTube in waves.
Open the transcript