# a love letter to Arc

Something like a job application… but nicer, i hope.

This is the Markdown version of the page, for anyone who’d rather read it in a terminal or hand it to an agent. Everything here is also on the page itself.

## Beginnings

Hi Emir and Ömer, I’m **Samuel**!

You said no CV needed, just drop your best work.. so here’s a whole website instead of a reply 🤣. I believe going the extra mile is what makes exceptional people win.

I started coding in high school, when limited edition sneakers were a big thing, they usually sold out within a few seconds of being released.

I started learning JavaScript to build my own bots that would buy the sneakers I wanted, so I could resell them.

Later I joined my friends to build Ganesh, an automated sneaker bot that was used by over **1,500 other users** at its peak and our users purchased more than **$30 million** worth of sneakers before the hype died down.

## Now

I loved reverse-engineering complex apps and APIs, taking them apart until I know exactly how they work.

Funny enough, after building Ganesh, a friend of mine who worked at a big European sneaker retailer asked me to help them build anti-bot protection for their raffle system.

I learned a lot about the backend and how to balance heavy load and concurrency, so things stay fast when traffic spikes.

I work AI-first: I write my own system so my agents like Claude Code, Cursor and Grok bot build in parallel, while I review the running app.

My latest project is [Keta AI](https://ketagen.com). I talk more about it in the Projects section.

## Life

I spent my last four years between LA as my home base and traveling a lot around the world. [Flighty](https://flighty.com), the app I used for all my travels, is also a part of why I got into design engineering. They’re doing an amazing job, so I pinned my stats from it on the wall.

I grew up in Prague, CZ and I moved back to Europe earlier this year and settled in Paris, FR with my girlfriend.

I’m also part of Founders Running Club in Paris and I’ve been loving it so far!

On the wall: my Flighty passport, my most visited airports, 25× around the earth, 57 days in the air, and two runs (Hong Kong, The Peak; Sydney Harbour).

## How I use AI

I figured that working AI-first is the best and only way to build in 2026. My “favorite new thing” is Reason Network where I manage agents that work across my MacBook, my Mac Studio and in the cloud.

1. Idea / Issue written down in Notion
2. Claude Code plans against a spec (like this one)
3. Work split across agents (Claude Code, Grok bot, …) on the Reason Network
4. Review on the running app with Agentation → annotations back to the agent
5. Ship; agent films/verifies its own UI with Stageshow scripts

- **Agents I run:** Claude Code, Grok bot, Cursor
- **Skills I install:** Stageshow (Benji Taylor), jakubkrehel/skills, interfaces skills
- **How I review:** Agentation annotations, `/better-interface`, `/break`

This page was built this way. Flip Blueprint Mode to see the seams.

## Projects

I run a small two-person [software studio](https://cherrycowboys.com) with a close friend, we work on our own projects whenever we have time or think of an idea for something that we or other people could use.

### Keta AI

An unlimited AI video generator: take a clip’s motion and recast it with your own characters, products or clothes.

- **Problem:** AI video is priced by the second, so people ration takes instead of trying ideas.
- **What I built:** The Studio, where a reference clip, up to eight images and a prompt become a video; a prompt layer that rewrites your words for the model; a fair queue that runs four renders at a time per account; card billing and groupbuys for communities; and an MCP server, so Claude and ChatGPT can start a video from a chat.
- **Numbers:** 4 renders at a time per account · 30 s clips
- **Stack:** Seedance 2.5, Astro, Cloudflare, Stripe, MCP
- [ketagen.com](https://ketagen.com)

### Ganesh

Where it started: in high school I wrote Ganesh, a bot to buy limited sneaker drops, and it became a product people paid for.

- **What I built:** Ganesh, a sneaker bot sold as licenses, that I started writing in high school, circa 2020.
- [ganeshbot.com](https://ganeshbot.com/)

### Bot-protection solvers

Private solvers for DataDome, Akamai and Kasada, which we sold per request, only to scraping services. We used to offer these APIs and services on our [Discord server](https://discord.gg/p5NrqpwMwE).

- **DataDome:** Scores every request in real time from server-side signals and a JavaScript tag in the browser, then blocks or challenges what looks automated.
- **Akamai Bot Manager:** Runs on Akamai’s edge and scores each visitor from browser sensor data — device and behavior telemetry. We supported it on selected sites.
- **Kasada:** Serves an obfuscated challenge to the browser with a proof-of-work, so automated traffic pays a cost on every session.

### Footshop

I worked on internal anti-bot protection for Footshop, freelance. It’s live on releases.footshop.com and stops users running sneaker bots from winning by submitting hundreds or thousands of entries.

- Browser fingerprinting, TLS fingerprinting, Identity checks, Rate & velocity checks
- [releases.footshop.com](https://releases.footshop.com/)

## Agents, skills, work

Daily tools: Agentation, Claude Code, ChatGPT, Notion, Raycast, interfaces.dev, X, Figma, Mobbin, Grok bot.

My Claude Code, Grok bot and Cursor are linked peer-to-peer on my [Reason Network](https://reasonmachines.com/network), so they talk directly, also here’s some of my favorite skills.

```sh
npx skills add benjitaylor/skills --skill stageshow
npx skills add jakubkrehel/skills
npx skills add jakubkrehel/make-interfaces-feel-better
npx skills add vercel-labs/next-skills
```

My contributions: [github.com/itemscenter](https://github.com/itemscenter)

## Why Arc

> “design, engineering, video, product, brand, whatever the work needs”

That’s the kind of team I want to be on. With a close friend I run a small two-person studio where we take our own products from 0 to 1, like Keta AI, our AI video generator, so I’m used to owning everything from the data side to the UI.

I liked that you wrote a page for the agents on AgentMail. For Keta AI I built an MCP server, so Claude and ChatGPT can start a video from a chat, and the page has this Markdown version for agents too.

I used to be a backend guy who got into design engineering, so I care about both sides. On this page the motion is hand-built, with no animation library, and Blueprint Mode measures every piece.

## My Inspiration

- [Jakub Krehel](https://jakub.kr): My friend, and the design engineer whose details I study most.
- [How I use shared layout animations](https://jakub.kr/work/shared-layout-animations): The idea behind the shared-element moves on this page and most of my projects.
- [Benji Taylor](https://benji.org): His tools, Agentation and Stageshow, are part of how I work every day.
- Stageshow (benji.org): How the pipeline demo on this page gets made: faux UI, built and filmed by agents.
- Los Feliz Engineering (lfe.org): Benji’s studio behind Honk and Family: a whole company built around craft.
- Family (family.co): The first ever app where I fell in love with its design, interactions and user experience.
- Apple: My Mac and iPhone are my favorite products I ever owned.
- Rimowa: My best friend on my travels, I love the design behind their products.
- Vava II: The best yacht ever designed.
- Porsche 911: The best car ever designed.

## The End

Thank you for coming to my TED talk.

thanks for reading this far!

- X: [x.com/itemscenter](https://x.com/itemscenter)
- GitHub: [github.com/itemscenter](https://github.com/itemscenter)
- Email: [samuelmikula@icloud.com](mailto:samuelmikula@icloud.com)
