A crypto project gets something wrong, so you go looking for help. Its website points you to a Discord. You open the Discord, and the pinned message warns you about impersonators and says staff will never DM you first. Then someone DMs you. They are in that same Discord, because the official support channel is also where the scammers wait. The official help path opens by telling you the official help path is dangerous.

I could not unsee that. So I built a company to fix it. Then I realized I was fixing it for the wrong customer.

More of normal life is moving on-chain every year. Not just trading. Savings first, and then, slowly, the things that sit behind savings: retirement, equities, the boring money. That shift brings a steady wave of people who are new to all of this. They are not here for the casino. They are people who happen to have real value on a blockchain now, and who will, eventually, need help with it.

Here is what help looks like for them today. The app or protocol sends you to its Discord or Telegram. The channel greets you with a scam warning about itself: do not trust DMs, staff will never message you first, verify every link. And the people that warning is about are right there in the channel, watching for someone confused enough to talk to. The support surface and the predator's hunting ground are the same room. Crypto-native users have fully normalized this. They learned the survival rules so long ago they forgot they ever had to learn them. A newcomer has not. The moment you say the loop out loud to someone outside crypto, it sounds insane, because it is.

The thing that first made me stop was small. A protocol's own site pointed to a Discord, and the Discord's pinned message was a scam warning about the Discord. The official channel, telling you the official channel is dangerous. Once I saw that pattern I started seeing it everywhere.

Then I watched large lending protocols start pulling out of Discord and Telegram, in part over exactly this. The support surface had become a liability they did not want to own.

So I built TxDesk. The first version was B2B: support infrastructure a protocol could run so its users got real answers instead of a scam-infested chat room. Underneath, the product did something specific, and it is the same thing it does today. It reads live on-chain data and explains it in plain language. It decodes a transaction, checks an approval, looks at what a contract actually is, and answers from live on-chain data. It is read-only with respect to your funds. It never asks for or holds a private key. The bet was that protocols would adopt this as their support layer.

Then I paid attention to what actually happens during an incident. When a protocol is being exploited, its attention goes exactly where it should: pause the contracts, trace the funds, pull in security people, coordinate, get a statement out. That is the correct priority for the protocol in that moment. The consequence is structural, not a moral failing. The thousands of individual users flooding in with "am I affected, was my wallet exposed, where are my funds, am I safe" are not, and cannot be, the priority in that window. The buyer I was selling to did not want to own that layer the way I had assumed. At the exact moment it mattered most, answering the individual user was nobody's job.

That reframed the whole thing for me. The people who have the questions, and the worry, and the money on the line, are the users. Not the protocols. The demand had been downstream the entire time. I had built the right tool and pointed it one customer too far upstream.

So I pointed TxDesk directly at the people who actually ask "am I safe." Same product underneath. Same thesis: people are coming on-chain and they need real answers, not a chat room with a scam warning pinned to the top. Different customer. The version I am building now lets the user ask the chain directly, in plain language, and get an answer grounded in what is actually on it.

Most people who follow my work knew TxDesk as a B2B product. I am not going to quietly pretend it was always this. It was not. I built it for one customer, learned the demand lived with another, and moved. The consumer version is what I am building now.