Aleksei

new account. i’m building my first startup in public and doing bug bounty + authorized pentesting. i’ll share real benchmarks, bugs, costs, product tests and failures. no fake mrr. no ai hype.

@debarghya_tech fair point. niche communities are probably the best place to start

@pentesti Reach niche communities where you ICP hangs out. Reddit is not the place at least at current times..

where i’d look for my first 100 users: - reddit for honest feedback - x for building an audience - linkedin for b2b leads - hacker news for technical users - product hunt when the product is ready you don’t need every channel. you need the one where your users already are.

@pulmencr “no privacy leaks” while your router watches you sleep is a wild sentence 😭

A guy just turned standard home Wi-Fi into a radar that sees through walls. Over 62,000 stars on GitHub for a system that tracks humans without using a single camera lens. It doesn’t just know you’re there. Watch what happens when a person lies down on a bed in another room. The AI instantly maps their exact posture, tracks their breathing, and calculates their heartbeat in real-time. No optical surveillance, no privacy leaks - it just analyzes how your body disrupts ordinary radio waves. The whole thing runs on a tiny $5 microchip. Traditional security requires a direct line of sight and invades your personal space. Now, your Wi-Fi router is a camera that sees through solid brick.

@shiri_shh MRR is a funny metric for something people hopefully buy once. this feels more like a premium wedding service with AI behind it than a SaaS.

$100K MRR app idea: couples upload two photos + how they met, get back a cinematic AI film of their love story across eras. weddings would pay anything for this

@fin465 the 90% reply rate is doing a lot of work here. the actual funnel is 300 texts → 10 demos: 3.3%. still solid, just a very different headline.

talked to a founder who hit $60k MRR at age 18 (yes its real, keep reading) he’s doing something i’ve never seen before with his outbound to get 90%+ reply rates: 1. targets realtors 2. builds a hyper-specific lead list + pulls their cell numbers in Origami 3. TEXTs them "hey is this [name] with [business]?" + throws in some personalization when possible 4. he gets ~290 responses off 300 texts 5. dump the repliers into an auto-dialer that calls 3 at once, pitch a demo live 6. of 50 that answer the phone, turns 10 into booked demos The 18 year old founder here is @adennurie_ (he's super impressive) And he's building all his lead lists with @origamichat -how we met. (more info in reply)

@chamath you can stop American companies from buying cheap intelligence. you can’t stop the rest of the world from using it. that’s not protection. it’s a self-imposed AI tax

The leading AI has already forked into two options. A: Closed source American that costs $26-56 per 1MM tokens. B: Open weight Chinese that costs $0.50-1 per 1MM tokens. If you force American companies to spend 50-100x more than their competitors abroad here is what will happen: 1. American companies spending 50-100x will at some point become financially impaired and the stock market will crater. This is the equivalent of the US Government saying we can only buy oil at $800/barrel even when the free market sells the same oil for $80/barrel. 2. In the short term, the revenues of the Closed American Labs may remain hi. But then their revenue will also crater because their customers in the US will be impaired and their customers abroad will have already flipped to cheaper solutions. This would be a terribly self-defeating form of intervention if it were to happen.

@jacob_posel the company brain will quickly discover that every team has a different answer to “who are we?”

How to become an AI native company: 1. Create a single MCP/API gateway. Let your team connect to your systems. 2. Create a company brain - Connect your static context ("Who are we", "guidelines", "product docs" ) - Connect your hot context (meeting notes, emails, slack, active projects) 3. Create a company harness. Instruct AI on how to interact with your company brain and systems. 4. Onboard your team into the harness. Create a self-improving loop. Let it learn from their work. 5. Create a model-routing layer. Evaluate and distribute work to the right model, at the right time. Remove vendor risk. 6. Build autonomous agents on top of your company harness.

@marclou the app was built in 2026. the review process is still living in 2008

1 afternoon to build an app 1 week to be reviewed

@thebuggeddev @Kimi_Moonshot Fable built the demo. Kimi built the project. suddenly those 3 hours don’t sound slow

My honest take on @Kimi_Moonshot K3 👀 As soon as I got access to Kimi, I put it through the exact same 3D football stadium challenge I had previously given Claude Fable 5. Fable 5 finished in under an hour. Kimi took almost 3 hours 🥵 At first, that was a very bad impression but i decided to see what's taking it this much of time. It was running E2E tests, validating desktop, tablet, and mobile, taking screenshots, finding failures, fixing them, even adapting for low end hardware and low FPS before moving on. All these that Fable 5 never did. None of it because it wasn't able to run this WebGL app on a headless browser. Kimi also shipped a clean React + Three.js codebase with proper components while Fable, as you might have checked codebase, generated one big HTML file with everything stuffed in. With all of this happening, it felt like I was doing more "Vibe Coding" with Fable 5 and more "Vibe Engineering" with Kimi K3. It feels like it's spending more time making sure the code actually works and easily scalable. Really impressed so far. More Kimi K3 experiments coming soon. Code (Kimi): github.com/thebuggeddev/f… Code (Fable): github.com/thebuggeddev/f… Live (Kimi hosted): stadiview.kimi.page

@z0rynx i built something similar for security research: 1,310 notes, 13,649 links, 403 sources. now my agents can search the vault, follow the links and pull the right context on their own. still testing it, but my notes finally feel useful.

A 27-year-old pointed Claude Code at 4,000 Obsidian notes and woke up to an employee he never hired. 6 years of notes. Ideas, half-finished essays, 212 book summaries all rotting in a folder he stopped opening in 2024. Then he noticed something obvious that almost nobody uses. Obsidian is just markdown files, and Claude Code lives in the terminal which means his entire second brain is one cd command away from an AI that can read all of it. So he ran it once, overnight. By 7 AM the agent had crawled every note, linked 340 orphan ideas back into the graph, flagged 18 contradictions between things he believed in 2021 and things he wrote in 2025, and drafted 6 essays from threads he forgot existed. Now the loop runs every night while he sleeps. New note goes in messy wakes up tagged, linked, connected to 3 older notes he'd never have found himself. His daily note greets him with what he was thinking exactly 1 year ago and why he was wrong. Cost of the setup: 1 folder path and a 400-word CLAUDE.md file. No plugins, no Notion AI subscription, no $30/month "second brain" course. People spent a decade building vaults that just sit there. He built one that thinks back.

@shengkun_ye if this survives X and LinkedIn changing the rules every other Tuesday, Apify might actually be in trouble

We just killed Apify. Your agent can now read every social media platform. No logins. No subscriptions. X, Reddit, LinkedIn, TikTok, Facebook, Instagram, YouTube, Rednote, and even Amazon. Apify: $199/month. Monid: From $0.0015/request. Pay as you go.

@psomkarbuilds Japan spent decades building intelligence with a body while the US built intelligence in a chat box. the current AI race mostly measures the second one

why is Japan so behind in AI they gave us robots bullet trains PlayStation and yet barely show up in the AI race what actually happened ?

@ErnestoSOFTWARE 70–80% margins are great, but chasing them too early is how you turn a growing business into a very profitable hobby

Making $2.5k from $10k is criminal you should be aiming for 70-80% profit margins This is how to profit max: 1. Do organic UGC not just ads. → Ads can get you 1.5 to 2x ROI → Organic can get you a 5 to 10X ROI do both, not just 1. 2. Make your onboarding shorter not longer shorter onboardings actually help to optimize your ads users that come from ads are already interested and don't need a 50 step onboarding. they just want to buy. This will help your CPA And find more qualified buyers 3. Have multiple retention offers through the app → you can use notifications → discount paywall → or free trials + discounts inside the app 4. Do everything you can to improve your product If your product is better your retention will be better And so will your ARPU ( average Rev per user) If you can make $2 per user Instead or $1 per user You're making 2x more money If you only look at upfront revenue and not at the important vitals of your app you will be left with criminally low margins on the long run

another Codex feature i almost missed: right-click a thread → Continue in new worktree same conversation, separate checkout. i use it for risky refactors. Codex can break a copy without touching mine. if it works, Hand off brings the chat + code back to Local. #codex

@levelsio realizing nobody knows what’s next is weirdly comforting

I increasingly have no idea about anything anymore, time is changing too fast, and with AI everyone is equal now, nobody really knows anything while there is a superintelligence that's actually moving things into a direction that nobody can predict

@YashHustle_22 Node.js. I’m too used to having TypeScript everywhere now

Which backend stack do you use the most? - Go - Node.js - Python - Java

@Dylan_txa_ the 6 ugly slides part got me. meanwhile i’m still fixing the landing page font.

I'm 30. I built an AI startup called GojiberryAI to $3.5M ARR. Got accepted into YC. If I had to start from 0, here's exactly what I'd do: 1. Sell it before I build it. No code. Just a simple slide deck (mine was 6 ugly slides) explaining the problem, the solution, the outcome, and the price. I made my first $10k that way, before writing a single line. 2. Pick a painfully specific customer. Not "B2B SaaS." Something like "founders at 20-person SaaS companies about to hire their first SDR." So specific that the right person reads it and thinks "that's me." 3. Start outbound on day one, but only to people showing intent. Not scraped lists. People engaging with competitors, changing roles, raising money, or publicly posting about the exact problem I solve. That's the gap between a 1-2% reply rate and 25-40%. 4. Lead with value, never a calendar link. Send a blueprint, not "got 15 minutes?" Let the resource do the selling, and the trial becomes the obvious next step instead of a pitch. 5. Pick ONE channel and go deep. For us it was outbound first, then Reddit (10M+ organic views), then LinkedIn lead magnets. I wouldn't touch a second channel until the first one was clearly working. 6. Talk to customers every single day. The product doesn't matter until you understand the problem better than they do. Spend 90% of every early call listening, not demoing. 7. Only build once people are actually paying. Then keep it dead simple and price it to sell itself. We landed on $99/mo with a free trial, so the funnel runs without me dragging anyone onto a call. 8. Do this relentlessly for about 12 months. That’s roughly how long $0 to $3.5M took us. Bootstrapped. No outside funding. Most founders don’t lose because they can’t build. They lose because they build too early, sell too late, and quit the channel before it compounds.

@YashHustle_22 TypeScript. i like javascript, I just dont trust myself with it

Which programming language do you prefer for web development? - TypeScript - JavaScript - Python - PHP - C# - Java - Other

found OpenShip, an open-source deployment platform you can run on your own server. it handles CI/CD, domains, SSL, databases, backups, logs and rollbacks. there’s also an MCP server, so Claude Code or Cursor can manage deployments. still early, but worth watching github.com/oblien/openship

@rauchg that would explain why everything breaks after an update

The big lesson from AI is that everything is code. A slide deck is code. Design is code. That cool promo video? Code. Excel automation? Code. The universe? Probably made of code too.

@levelsio apps may not disappear. we might just stop opening them and let agents use them for us

Also some thing I forgot to add There's so many things now where I don't even use the app anymore but just generate my own app to do things Like WHOOP, for years I wanted to know the correlation between stuff I do and my workouts and sleep, like going to sauna, or what I eat. WHOOP has a journal but it's practically unusable, I logged sauna for ahwile, then I realized you have to ALSO log the days you DO NOT go to sauna, by hand, every day! Useless So in those cases I just ask Claude Code to build a little web app for me that pulls all my data in from my WHOOP, and what I eat, my @wip logs when I go to sauna, or go tan, or go gym etc. It just pulls all the data in and gives me the answers to questions I have Half the time I don't even use the web app it generates, I just ask it directly stuff I have to find a hotel now for some trips we go, I didn't even go to my own site Hotelist, I just asked Claude Code on my Hotelist VPS server to find me a hotel, etc. So many times where it just saves more time to ask Claude Code directly stuff than try to do it with some startup/company's app That's what I mean, I'm obviously more tech than regular people, but we usually slightly ahead in how we use tools, and if I stop using websites and apps, and just use Claude Code for everything in my life, it's not without logic to think that something like that will be the future for most people And of course it's already happening with Claude Cowork and Codex app which regular non-tech people use My point is that the entire app and website layer is just getting decimated And me being a person who lived kinda in that layer (well more websites ofc to be honest) means I have no clue where it's going to except this, or what a person should do, for the first time, I have no clue The only thing I know is that the superintelligence clanker can't reach into IRL and hardware yet for the next few years at least, and there's lots of stuff there to do that's differentiating, because you need a space, tools, machines, make things, and no you can't generate that in 3 minutes with an AI Blood, sweat and tears are not automated yet by AI, yet!!!!