Scam on you

Denis Perov — Fri, 21 Nov 2025 08:30:04 +0000

I just wasted scammers' time by pretending to fall for their LinkedIn job offer.

Few days ago got message from a stranger account, 500+ connections, employee of EOS Venezuela Block Producers. Company page looked real with 80+ followers, California headquarters, multiple employees listed.

But something felt wrong from start. My curiosity won. Decided to play along.

After I shared my resume, they asked for a test assignment. Connected me with another person via Teams who shared "Node application" source code and asked to run it. Long story short, there was a a malicious code download script hidden in tasks.json for VSCode.

Don't trust outsiders, challenge their statements:

Take time to think before doing anything
If someone too excited to give you job this is a 🚩
Never run code from strangers

Tickets that don't crash when it matters most

Denis Perov — Tue, 04 Nov 2025 11:02:27 +0000

💃 25,000 Jennifer Lopez tickets. 4 hours of complete system failure. People were charged for tickets that never existed. I spent 1 week building a blockchain solution with parallel execution, reaching 10k+ TPS.

TL;DR: Project almost died midway. Blockchain broke. Thought that I had to fake the demo, but got it working the day before the deadline.

What can make AI companies pay for data instead of scraping?

Denis Perov — Wed, 27 Aug 2025 20:07:59 +0000

We all know the current state of AI data scraping is broken. AI companies are consuming massive amounts of bandwidth (seen cases of 30TB/month from single bots), driving up infrastructure costs, and training on content without compensating creators. Meanwhile, content creators are stuck either blocking everything or watching their servers burn 🔥

There's Cloudflare's pay-per-crawl (when it releases) that just adds paywalls to web scraping which is essentially a cat-and-mouse game because 30-40% of AI traffic is now unidentified bots.

We need a completely different solution, which will motivate AI companies to pay for the content. And to achieve that, we must give them a better alternative to web scraping 💡

The Concept

I though of a different approach -- a data marketplace where AI companies pay content creators directly for structured, clean training data.

Instead of the current chaos of scraping, blocking, and legal battles, this creates a professional data exchange where:

Content creators export their data to external data storage
AI companies access structured data via MCP server using SQL-like queries
Data is split into two chunks: public & private. Public chunk can be used for filtering but access to the private chunk is token-gated
Smart contracts handle automatic payments in cryptocurrency and access control
Everything is permissionless -- no middleman taking cuts or controlling access

This approach offers:

Structured data that AI companies actually prefer over HTML scraping
Zero infrastructure burden - data served from decentralized network
Cryptographic proof of licensing for auditing

Development Plan

Planning to build a WordPress plugin as the first integration because:

WordPress powers 43% of the web - massive potential impact
Site owners are desperately seeking solutions (27+ blocking plugins exist)
Non-technical users need simple tools to participate

The plugin would provide simple admin UI for data export configuration, public/private field mapping (e.g., titles free, full content paid), automatic sync of new content and earnings dashboard.

Seeking Feedback

This isn't about forcing anyone to monetize their data -- it's about giving creators the option to get compensated instead of just bearing the costs. If you want your data free for AI training, set the price to zero. Your choice.

Happy to dive into technical details. Also interested in connecting with WordPress site owners who want to "make money from the machines" and might want to beta test.