Mihaylo Petruk
Offer Mihaylo work on your next project.
Private message
Offer work
Kyiv,
Ukraine
3 days 5 hours back
available for hire
age 19 years
on the service 5 years
Rating
Successful projects
Average rating
Language proficiency level
Українська: fluent
Русский: fluent
English: upper-intermediate
CV
Machine learning specialist with six months of experience in developing object detection models. I have practical experience in data preparation, tuning and optimization of models, in particular using TensorFlow and PyTorch. I am confident in data augmentation methods, have worked with YOLO and Faster R-CNN architectures. I focus on improving the accuracy and performance of real-time object recognition models. I also have experience and skills in design development (3d-max blender, Photoshop)
Skills and abilities
Programming
Services
Portfolio
-
226 USD Neural model for detecting enemy military equipment
A deep neural network model has been developed for automatic detection of enemy military equipment in images and aerial videos (drones, satellites, surveillance cameras).
The model identifies types of equipment (tanks, APCs, artillery, trucks, air defense, etc.), and can be integrated into combat or reconnaissance systems.
-
113 USD Trained model for drone (UAV) detection
Developed a complete system for automatic detection of drones in the airspace on videos and images using computer vision and deep learning methods.
The system can be used for object security, real-time drone detection, and integration with tracking or air defense systems.
-
68 USD Neural network for recognizing different types of insects
AI & Machine Learning
This project is aimed at creating a system based on neural networks for automatic insect recognition by their type. The main goal is to provide accurate insect classification for use in the agricultural sector, environmental monitoring, or scientific research.
-
68 USD Neural network for recognizing various types of insects
AI & Machine Learning
This project is aimed at creating a system based on neural networks for the automatic recognition of insects by their type. The main goal is to provide accurate classification of insects for use in the agricultural sector, environmental monitoring, or scientific research. The system is capable of operating with high accuracy even in field conditions, allowing for the automation of the complex process of analyzing biological material.
Activity
| Latest proposals 10 | Budget | Added | Deadlines | Proposal | |
|---|---|---|---|---|---|
|
Create a Telegram bot
271 USD
|
|||||
|
Set up recording of a separate window.
17 USD
|
|||||
|
Telegram bot service desk for a network of bakeries (15+ locations) with Google Sheets integration
338 USD
|
|||||
|
Sparsity catalog of auto goods
23 USD
|
|||||
|
Bot parser of advertisements from necessary websites by parameters
113 USD
|
|||||
|
Telegram bot
34 USD
|
|||||
|
Counting of finished products and people involved in the process based on the YOLO model.
|
|||||
|
TG script for mass invitation
23 USD
|
|||||
|
Audience targeting parser in Telegram
23 USD
|
|||||
|
Monitoring prices on Tabletka.ua
406 USD
|
