.jpg/:/rs=w:1353,h:391)
Hi. I'm Andrii Volkovskyi
I'm physicist (specializing in solar cells and semiconductor nanofabrication)
and Junior AI Engineer (specializing).
My education
2016 - Master degree in Physics (specialization Photonics)
at Taras Shevchenko National University of Kyiv (it's in top 10 universities in Ukraine)
2016 - Master degree in Renewable Energy Science and Technology
at Ecole Polytechnique, France (it's in top 50 universities in the world)
2024 - completed several CS50 courses (CS50x - Introduction to Computer Science, CS50P - Introduction to programming with Python, CS50AI - Introduction to Artificial Intelligence with Python)
at Harvard Online, CS50
Experience
Jun-Aug 2012: worked as Process Manager at Icicle Seafood
It was summer work during Work and Travel program in USA for students
Mar -Jul 2015: worked as Intern at Laboratoire d'Optique Appliquée UMR7639 CNRS
It was internship in Laser-Plasma Acceleration project
Mar-Aug 2016: worked as Intern at Électricité de France
It was internship in perovskite sollar cells project
Mar 2017 - Dec 2020: worked as part of Scientific staff and PhD candidate at Walter Schottky Institut, Technische Universität München
It was PhD program connected to semiconductor nanofabrication (Photonic Crystal Cavities and TMD Materials) and optical characterization. I also had Doctoral Fellowship from Marie Skłodowska-Curie Actions
Sep 2021 - present: work as private tutor of Math and Physics
When I returned to Ukraine after my postgraduate studies, while I was looking for a job, I decided to work as a tutor in physics and mathematics. However, then a full-scale war began, and I could no longer go abroad for work. In Ukraine, there are no jobs in my field because we don’t have factories for semiconductor chip production. So, I continued working as a tutor and simultaneously studied programming.
My projects
Masked word prediction
During the CS50AI course, I had to write an AI model that can predict the missing word in the sentence using attention mechanism and transformers library.
Traffic Signs Recognition
Another project I completed during the CS50AI course involved writing a convolutional neural network that can recognize and distinguish road signs.
Gender Recognition
I made a convolutional neural network which can take images of people and predict their gender. It was trained on a public set of images of different celebrities.
Python 3D shooter game with AI NPCs
I took engine of Doom written in Python and modified it into my own mod, which I named "Doom - Kursk People’s Republic". There is also a Machine Learning part in the code, where bots use breadth-first search to find the player on the map. The game was made as a joke just for fun.
License Plates Detector with YOLOv8
The goal of this project is to create a license plate detector. I took a pre-trained YOLOv8 model, and fine-tuned it on a public dataset of car images. The final output is a model that can be used to identify license plates and it was tested on real-time video streams.
AI vs Human Detector with RoBERTa
The goal of this project is to create the detector which can classify text written by AI and by Human. I took a pre-trained RoBERTa model, and fine-tuned it on a public dataset of different texts written by human and AI. The final output is a model that can be used to identify generated text but it has not very good accuracy.
Face Detection and Blur
A simple Python application that uses a graphical interface (Tkinter) to detect or blur faces in images, videos, or live webcam feeds with OpenCV and MediaPipe.
AI Multi-agent Grocery Shopping Assistant
The multi-agent AI assistant which can propose the recipe of the meal and plan a shopping list for that meal. It uses LangGraph, include logging via LangSmith, leverage LLMs through LangChain, and provide a simple UI using Streamlit and speech recognition using Whisper. It also has API integration with SerpAPI to get real prices from Walmart.
My Hobby
I also have a hobby. That's my youtube channel about food and cooking in Ukrainian language.