I build products used by millions
I'm Yurii Liubchak, AI lead at KNESS. I build production AI systems and run Telegram products used by millions of people.
What I'm doing
Main workAt KNESS Group I lead AI and automation work. I build internal AI systems around company data and tools: document analysis, internal search, employee workflows and department-specific agents for HR, Legal and Accounting. I work on them from prototype to production and keep them running.
Independent workI run fStik, QuotLy and other Telegram products on my own infrastructure, and build Capka, an open-source AI workspace. I also occasionally work with companies on AI and automation.
Selected work
-
KNESS · AI enterprise · production
Internal AI systems used in company workflows across HR, Legal, Accounting and other teams.
-
fStik fstik.app · @fStikBot
The largest sticker platform on Telegram, 800K monthly users. Bot, site, Mini App and an Android app on one stack, built and maintained solo
-
Capka open source · self-hosted
An open-source AI workspace where every chat gets its own Linux sandbox and file storage. Agents can run code, work with files and return finished artifacts.
-
QuotLy @QuotLyBot
Turns any Telegram message into a clean quote image, 80K groups, up to 48 000 messages per minute at peak
More products, experiments and notes · @LyBlog →
About
Started at twelve, editing uCoz sites with too much time on my hands. Then came PHP, then VK bots, then the Telegram ecosystem I still live inside today.
Most of my public products grew out of Telegram. fStik and QuotLy together reached millions of users, and in 2023 I took first place in Telegram's Mini App Contest.
These days most of my work is around AI, infrastructure and automation, both at KNESS and in my own projects.
I still like building things myself and running them in production. I write about it at @LyBlog.
How I build
TypeScript / Node.js for most backend work, React / Next.js on the frontend, Python where it makes sense. Telegram through Telegraf, grammY, TDLib, MTProto and Mini Apps. Infrastructure is mostly Linux and Docker on my own boxes, with Postgres, MongoDB, Redis and Sentry
Get in touch
Working on an AI system, automation or a product that has to handle real scale? Tell me what's broken.
- Email[email protected]
- Telegram@LyDev
- GitHub@LyoSU
- LinkedIn/in/lyubchak
- Blog@LyBlog