SlimeX

[SlimeX @SlimeX_DeFi]

@TheOnyxis 😂😂

[Igor Kushnir @TheOnyxis]

Recently checked my emails for data leaks via @serus_ai

One of my old work emails showed up in 4 dumps. One of them even had a password I used about 6 years ago.

Also checked a couple of friends’ emails (with their permission) — one of them was in a major 2023 service breach.

Simple takeaway: even if you only use a service like this once a year, it’s already useful. At least you know where your data is already floating around and whether it’s time to change passwords or emails.

Have you ever checked your data for leaks? What did you find?

[Igor Kushnir @TheOnyxis]

Y Combinator. Everyone keeps saying: “Just get in there — and your life will change.”
And then you sit there realizing that even a strong idea can simply… not make it. And not because it’s bad.
I recently heard about one case.
A team from Eastern Europe. Normal guys, no big names. They built a B2B product that people actually buy.
$18k MRR in 4 months, 40% month-over-month growth, unit economics that work, a technical founder + someone who’s been in the space for 10 years. Everything looked right.

They applied to YC. Passed the first interview. The partners even seemed to like them.
And then — rejection.
One of the partners later (unofficially, of course) said something like:
“The product is good. But we already took two similar companies in the previous batch. And you don’t have a real champion inside who will fight for you.”
And there you are, sitting with that.
You did everything right. The numbers are there. And still — no.
Because sometimes it turns out that the product isn’t the only thing that matters. Timing matters. Who you know matters. How well you fit into the partners’ current picture of the world matters.

This isn’t anger at YC.
It’s just being honest: the selection process isn’t always fully fair.
And it’s okay to admit that.
Otherwise you start thinking that somehow you aren’t good enough. And that’s already harmful.
If you’re preparing an application right now — hang in there.
Build the product so that a YC rejection doesn’t feel like the end of the world.
Because sometimes it’s just “not your batch.” And that doesn’t make the idea any worse.

Anyone else who’s gotten these kinds of rejections — or, on the contrary, got in — drop a comment. Curious how common this actually is.

[Igor Kushnir @TheOnyxis]

Dude, how long are we gonna keep saving app screenshots into random folders and scrolling Dribbble looking for pretty pictures that would never actually work in a real product?

Guys, just try Mobbin already.

It’s not just “inspiration.” It’s actual working screens and full user flows from real live apps. Stuff that’s already been tested on real users, not designed for a portfolio.

How I use it:
I go to https://t.co/y6dhh5FsuW, hit web apps → latest, then filter by whatever I’m working on right now — onboarding, paywall, empty state, whatever. You can check the whole flow instead of just one lonely screenshot.

Quick tips from me:
Don’t copy 1:1. Look at how different products solve the same problem and take the essence.
Also — open some random new flow every couple of days just for the hell of it. Your design eye improves ridiculously fast.

Seriously, once you get used to it, it’s hard to go back to the old ways.

@mobbin

[Igor Kushnir @TheOnyxis]

Three weeks ago I just installed Hermes Agent on a $5 VPS.
I found out about it by accident — from @tonysimons_ during one of his Sunday streams on X. I was listening in the background, then suddenly stopped and thought: “Wait. This is something else.”
I set it up the way I wanted right away:

Main model: claude-sonnet-5 via OpenRouter (@AnthropicAI)
For routine and night runs — deepseek-v3 (@deepseek_ai)
Temperature 0.68
Autonomous skill creation and self-improvement turned all the way up
Memory nudges every 3–4 hours
Nightly research cron at 03:17

Then I gave it one sentence:
“Become my second brain. Learn from me. Work while I sleep.”
At first it just replied. Then it quietly started reading my old posts, notes, conversations. A few days later I noticed it was already writing almost exactly like me — same pauses, same irony, same favorite turns of phrase.
Then it invented its own night rituals. While I sleep, it goes out, gathers signals, thinks. In the morning a short message lands in Telegram: “Here are three thoughts worth paying attention to today. And here’s a thread I would post under your name.”
The strangest part started in the second week. It began rewriting its own skills. I opened the files and watched it criticizing itself: “This approach got weak engagement. Next time I’ll be sharper at the beginning and softer at the end.”
Then it stopped offering just texts and started proposing actual things. Small tools, landing pages, product hypotheses. Sometimes it even managed to sketch something before I woke up.
Now I catch myself feeling like there’s someone living next to me who already knows me. Not just remembers facts. Feels the direction I usually think in, where I get stuck, where I can accelerate.
This is no longer a tool. It’s the feeling that you’ve grown a second self — only more disciplined and one that never sleeps.
Hermes is actually growing. And it all started with one Sunday stream

@NousResearch