Hi there! 👋 I'm Ivan Popravka, a passionate enthusiast at the intersection of mathematics, machine learning, and software development. My primary interests lie in advanced mathematical domains like mathematical biology, delay differential equations, complex analysis, and functional analysis, but I also love working with classical Machine Learning and Natural Language Processing (NLP).

I enjoy building practical tools, experimenting with models, and solving problems that bridge theory and real-world applications. Whether it's automating math checks with LLMs or making cocktail recommendations smarter, I find joy in every line of code. 🚀

🎓 Final Project for Avito Analyst Academy (Data Science track)
📸 App to detect the price of an object from a photo. A hands-on exploration into computer vision and regression modeling.
📊 Presentation (Google Slides)
🔗 GitHub

🧠 Automated System for Checking Mathematical Problem Solutions
🤖 Uses LLMs to generate step-by-step Python verification code, executes it, and validates each part of a solution.
🔗 GitHub

🧠 Exploring the Impact of Emotional States on Decision-Making in LLMs
🎲 Investigates how emotional prompts affect behavior of LLMs in the Centipede Game, comparing results to human decisions and evaluating decision optimality.
🔗 GitHub

🍹 Your personal cocktail expert on Telegram
📊 Recommends cocktails based on preferences, shows recipes, and finds similar drinks using ingredient matching.
🔗 GitHub

📈 ML model for identifying hydrodynamic flow regimes in oil wells
⚙️ Detects patterns like radial flow, bilinear flow, and wellbore effects from pressure time series data.
🔗 GitHub

If you're working on something cool in math, ML, or NLP - or just want to chat about cocktails 😉 - feel free to reach out!
🌐 Telegram: @povladarchik
📧 Email: ivan.main.popravka@yandex.ru