Sneaker bot
Automated checkout for hype drops: 200+ pairs of Yeezy 350 Zebras in one run.
- Role
- Builder & operator
- Stack
- PythonPhantomJSPuppeteerProxies
Where it started. From 2016 I built and ran a bot for hype retail releases on Yeezy, Adidas, Nike and Supreme — drops where thousands of people want a few hundred pairs, and they're gone in seconds.
I sold it through hypebeast forums and local sneakerheads, either running it for them as a service or selling them a copy to run themselves. It made $50K+ in personal revenue and high six figures for clients.
Decisions
Milliseconds decide the outcome
On a hype release, the moment a regular person opened the page, it was already over. Everything came down to how fast a checkout hit the store once a drop went live.
It's the problem I still solve in trading: the result is decided before a human can react, so the system has to be ready first.
Skip the storefront
When retailers moved drops behind waiting rooms and queues, the bot stopped shopping like a person. It worked directly with the purchase flow and ran sessions in parallel so one always got through, keeping pace with most of the defenses of its time.
What I built
- Written mainly in Python, driving headless browsers: PhantomJS at first, then Puppeteer.
- Pools of purchased IPs, so many sessions could run at once without looking like one buyer.
- Saved payment and shipping profiles, so a checkout could be submitted the instant a product went live.
- Best run: over 200 pairs of Yeezy 350 "Zebras" for a client who owned a resale store, worth $200K+ on the secondary market at the time.