> ls -la /whoami

Websites, Telegram services and AI tools for business

I turn ideas into working products: from a landing page or bot to an internal service with integrations and AI features.

View cases What you can order Discuss a project

2006 started the path

100+ projects

∞ lines of code

LLM researching

Code is a way of thinking, not just a profession.

/services

What you can order

Choose a familiar type of task — then we will define the scope, timing and shortest route to launch together.

Websites and web products

Landing pages, personal portfolios, client portals and services with a thoughtful interface and reliable backend.

View similar cases ↗ 02

Telegram bots and Mini Apps

Bots, mini-apps, payments, notifications and an admin panel — a complete customer flow inside Telegram.

View similar cases ↗ 03

Business automation

Internal tools, integrations, operational workflows, reports and routine processes that should not live in spreadsheets.

View similar cases ↗ 04

AI features and assistants

RAG, support assistants, document processing and AI features with sources, evaluation and human control where it matters.

View similar cases ↗

/projects

Cases from real work

all telegram web products automation ai desktop

BotHub: Telegram mini-app marketplace

Laravel 12 React Telegram Bot API Telegram Mini App MoonShine MySQL Queue Payments

A Telegram-oriented platform that combines link/bot aggregation, storefront, and checkout in one product without a separate registration flow.

↗

MoonShift: shift and payroll operations in Telegram

Laravel 12 MoonShine Telegram Bot API Matrix Scheduler Queue MySQL

A management system for dispatch teams where operators work with Telegram bots and administrators use one control panel for schedules, attendance checks, payroll, and reports.

↗

Support AI assistant: retrieval + moderation layer

LLM RAG Python OpenAI Moderation Queue PostgreSQL Observability

A support workflow that blends retrieval, safe-prompt constraints, and analytics to reduce repetitive support load in multilingual products.

↗

> view all projects →

/timeline

20 years of practice: from first websites to an AI/LLM lab.

2006 Started building websites as a hobby

2008 University and first commercial projects

2012 Desktop development and complex systems

2016 Web, backend, frontend and architecture

2023 LLM, AI and new horizons

2026 Still learning and building new things...

still building →

/lab

Experiments and research

The lab is a working bench for LLM, RAG, AI planning, code intelligence and admin workflows. I document testable hypotheses, not shiny promises: what can be automated, where a human must stay in control and how to bring it to production.

> llm-playground

in progress

A shared bench for quick checks: prompts, system instructions, tool calling, JSON responses, streaming and model behavior on the same inputs.

ai-agent code-miner prompt-studio rag-lab

> run experiment.py loading model... gpt-4o building context... ok generating... ███████████ 78% result: useful

// principle
I am not looking for magic tools. I assemble working systems.

/notes research log

Field notes from the AI lab

LIVE BUILD NOTES

RAG pipeline without magic

How I assemble a retrieval layer: sources, chunking, pgvector, eval sets, tracing and honest quality checks.

lab: rag-lab / retrieval / evals

AI Factory task planning loop

How a planner should work: goal, decomposition, skill selection, checkpoints, execution trace and manual approval for risky actions.

lab: ai-factory-planner / skills / execution trace

Context mining for large codebases

How to extract not just files from a repository, but a useful context pack: entrypoints, dependencies, risk zones and decision history.

lab: code-miner / repo context / LLM handoff

/about :: senior fullstack operator

I build products where interface, data and AI model have to work as one system.

I cover the full cycle: architecture, prototype, production infrastructure, observability and careful UX. I like complex domains where a hypothesis needs to be tested quickly without losing engineering discipline.

TypeScript React Next.js Node.js Python Postgres Redis OpenAI API RAG Queues CI/CD Telemetry

INPUTcontext.md

VECTORpgvector index

MODELreasoning loop

TOOLSapi + workers

$ run senior-stack --mode=production
architecture: clean | telemetry: on | ai: assisted
output: systems that survive real users

/contact :: open channel

Have a task? Let’s work out how to launch it.

Write about the idea, current product or process you want to automate. I will help formulate a solution and honestly assess the next step.