Skip to content
View isad4youbro's full-sized avatar
😑
😑

Block or report isad4youbro

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
isad4youbro/README.md

Art

ML / AI Developer · Python · Tabular ML · Recommender Systems

11th-grade student from Saint Petersburg building practical ML systems and end-to-end prototypes.

GitHub Projects


🇬🇧 English

About

I build machine-learning systems around tabular data, text classification, recommendation and personalization.

My usual workflow is not just training a model: I try to take a solution from data → features → validation → inference → working service.

I am especially interested in practical ML with Python and gradient boosting, and in understanding how a model becomes part of a real application.

🏆 Highlights

Result Competition / Project What I worked on
🏆 Finalist PROD 2025 — T-Bank × HSE × Central University Personalized cashback with LightGBM
🥇 1st place AI Business SPB / MediaScope Search-query classification and title recovery
🥇 1st place My Profession IT — regional stage Recommender / ML solution
🥉 Top 3 Smart Petersburg — regional hackathon ML / software project
🏆 Winner PROD — first selection stage Industrial development olympiad

🔬 Featured projects

End-to-end ML service for personalized cashback.

LightGBM · scikit-learn · pandas · FastAPI · Docker

A portfolio-safe version of a competition project. The service combines model scoring, feature construction and business constraints, then exposes the result through a documented API.

Highlights: ML predictor · category ranking · fallback heuristics · request logging · health checks · Swagger / OpenAPI.

Hybrid ML pipeline for video-search query classification and title recovery.

TF-IDF · Logistic Regression · LinearSVC · KNN · GroupKFold

The solution combines text models, retrieval, dictionaries, heuristics and domain-specific post-processing instead of relying on a single classifier.

Telegram automation project with scheduled jobs and persistent storage.

Python · Cloudflare Workers · D1 · Telegram API · CRON

A backend-oriented project demonstrating bot logic, scheduled execution, storage and deployment-oriented architecture.

🧰 Tech stack

Languages

Python · SQL

Machine Learning

LightGBM · CatBoost · scikit-learn · pandas · NumPy · TF-IDF

ML skills

Classification · Regression · Recommender Systems · Feature Engineering · EDA · Model Validation

Backend / Engineering

FastAPI · Pydantic · Docker · OpenAPI / Swagger · Git

Tools

Jupyter · Streamlit

🎯 What I am looking for

I am interested in ML / AI internships and practical projects where I can work with real data, experiment with models, learn from experienced engineers and turn ideas into working systems.

📫 Contacts


🇷🇺 Русская версия

Обо мне

Я занимаюсь машинным обучением, табличными данными, классификацией текста, рекомендательными системами и персонализацией.

Мне интересно не только обучить модель, но и довести решение до работающего результата по цепочке:

данные → признаки → валидация → инференс → сервис.

Основной стек — Python и классическое ML, особенно градиентный бустинг. Отдельно интересуюсь тем, как ML-модель встраивается в полноценное приложение.

🏆 Достижения

Результат Соревнование / проект Что делал
🏆 Финалист PROD 2025 — Т-Банк × ВШЭ × ЦУ Персональный кэшбэк на LightGBM
🥇 1 место AI Business SPB / MediaScope Классификация поисковых запросов и восстановление названий
🥇 1 место Моя профессия ИТ — региональный этап ML / рекомендательное решение
🥉 Топ-3 Умный Петербург — региональный хакатон ML / программный проект
🏆 Победитель PROD — первый отборочный этап Олимпиада по промышленной разработке

🔬 Основные проекты

Полноценный ML-сервис персонального кэшбэка.

LightGBM · scikit-learn · pandas · FastAPI · Docker

Публичная версия соревновательного проекта. Сервис объединяет ML-скоринг, построение признаков и бизнес-ограничения и предоставляет результат через документированный API.

Внутри: ML-предиктор · ранжирование категорий · fallback-эвристики · логирование запросов · healthcheck · Swagger / OpenAPI.

Гибридный ML-пайплайн для классификации видеопоисковых запросов и восстановления названий.

TF-IDF · Logistic Regression · LinearSVC · KNN · GroupKFold

Решение объединяет текстовые модели, retrieval, словари, эвристики и предметные post-processing правила вместо использования одного классификатора.

Backend-ориентированный Telegram-проект с планировщиком задач и постоянным хранилищем.

Python · Cloudflare Workers · D1 · Telegram API · CRON

Проект показывает работу с логикой бота, scheduled jobs, хранилищем и deployment-oriented архитектурой.

🧰 Технологии

Языки: Python · SQL

ML: LightGBM · CatBoost · scikit-learn · pandas · NumPy · TF-IDF

Задачи: классификация · регрессия · рекомендательные системы · feature engineering · EDA · валидация моделей

Backend / Engineering: FastAPI · Pydantic · Docker · OpenAPI / Swagger · Git

Инструменты: Jupyter · Streamlit

🎯 Что ищу

Ищу стажировки и практические проекты в ML / AI, где можно работать с реальными данными, экспериментировать с моделями, учиться у сильных инженеров и превращать решения в работающие системы.

📫 Контакты

Pinned Loading

  1. cashback-ml cashback-ml Public

    ML service for personalized cashback offers with budget-aware ranking and FastAPI inference API.

    Python 1

  2. mediascope-query-classifier mediascope-query-classifier Public

    Hybrid recommendation and query classification pipeline for video content using ML and rule-based post-processing.

    Python