Custom
developmentFrom 2-week bots to production ML. Laravel and Next.js sites, parsers, integrations — remote, contract-based, fixed price once the spec is agreed.
Telegram bots in Python
Bots that handle the boring stuff: notifications, reminders, channel parsing, mini-apps with payments and admin panels. Built on aiogram 3 — the de-facto standard for scalable Telegram bots.
- Helper bots: reminders, checklists, FAQ
- Catalog bots with search, cart, payments (Stripe, CryptoBot)
- Channel parsers and scheduled auto-posting
- Telegram Mini Apps: Next.js UI inside the bot
- Integrations with CRMs, Google Sheets, databases
Laravel websites
Backend-heavy projects where admin panels, roles, queues, and integrations matter. Laravel has been my main stack — Filament and Livewire are second nature, so a new CRUD module takes a day, not a week.
- Corporate sites with Filament-powered CMS
- User dashboards and SaaS with subscriptions (Stripe, Paddle)
- Marketplaces and catalogs with filters, search, analytics
- REST and GraphQL APIs for mobile apps
- High-load setups: Redis, queues, horizontal scaling
Next.js web applications
When you need a fast UI, SSR for SEO, and the React ecosystem — Next.js. The portfolio site you are looking at runs on it: App Router, Server Components, ISR, next-intl. Good fit for marketing sites, dashboards, and AI interfaces.
- Marketing sites with SSR and clean Lighthouse scores
- Dashboards and admin panels with realtime updates
- AI interfaces: chats, generators, assistants
- Multilingual sites with proper hreflang
- Headless CMS integration (Sanity, Strapi, Payload)
ML models and data analysis
Applied ML: classification, recommendations, NLP, tabular data. My approach — ship a simple baseline first, then improve. No LLM hype where logistic regression solves the task.
- Classification and regression on tabular data
- Recommender systems with collaborative filtering
- NLP: text classification, entity extraction, topic modeling
- Dataset analysis with insight reports
- Model deployment as a FastAPI REST API
| age | chol | thalach | oldpeak | target | |
|---|---|---|---|---|---|
| mean | 54.4 | 246.7 | 149.6 | 1.04 | 0.54 |
| std | 9.1 | 51.7 | 22.9 | 1.16 | 0.50 |
| min | 29.0 | 126.0 | 71.0 | 0.00 | 0.00 |
| max | 77.0 | 564.0 | 202.0 | 6.20 | 1.00 |
Parsers, integrations, automation
The most common request: pull data from a site, push it into a CRM, notify in Telegram. Quick to deliver — most building blocks are ready, the work is gluing them to your case.
- Web scraping with bypass for basic protection (Playwright, proxy)
- Marketplace integrations: Amazon, Etsy, eBay
- Google Sheets ↔ CRM ↔ database sync
- Event-driven email and Telegram notifications
- FastAPI or n8n workers for repeating tasks
How I work
A transparent process with no surprises: from the first call to delivery and support.
Free 20–30 min call
We discuss the task, I ask clarifying questions, give a rough timeline and price range.
Spec and fixed quote
I write a short spec. Once it is agreed, I commit to a fixed price and deadline.
1–2 week sprints
Each sprint ends with a working demo. You see progress and can adjust along the way.
Handover and support
Deployment, docs, walkthrough, 30 days of free bug fixes. After that — hourly or fixed-scope work.
Common questions
Want to discuss
your project?
Free 20–30 min intro call with a rough estimate, no commitment. I reply within the day.