Serhii Savenkov
Offer Serhii work on your next project.
Private message 
Offer work 
Dnepr,
Ukraine
1 day 4 hours back
available for hire
1 proposal made
on the service 3 years
Rating
Successful projects
Average rating
Language proficiency level
Русский: fluent
Українська: advanced
Skills and abilities
Programming
Mobile development
Portfolio
- All works 14
- Javascript and Typescript 1
- Gaming Apps 5
- AI & Machine Learning 5
- Cryptocurrency & Blockchain 1
-
Strange Clan
Gaming Apps
Strange Clan is an online, action RPG, using Play-2Own mechanics to give players ownership of their digital items in game.
-
Video Demo person detection for re-identification
AI & Machine Learning
Technologies of computer vision for person detection and person re-identification based on C++, OpenCV, Caffe etc. Surveillance video analysis
-
BBQ Coin
Gaming Apps
BBQ Coin Tap is an engaging and interactive miniapp available on Telegram that allows users to earn virtual BBQ coins by tapping.
Simple and Fun Gameplay:
Users tap the screen to earn BBQ coins.
… The faster and more frequently they tap, the more coins they collect. The intuitive interface makes it easy for users of all ages to enjoy.
Rewards and Incentives:
Accumulate BBQ coins to unlock various rewards and bonuses.
Compete with friends or other users to achieve high scores and leaderboard positions. Special events and challenges provide additional opportunities to earn extra coins.
Integration with Telegram:
Easily accessible within Telegram chats and groups.
Share achievements and invite friends to join the fun directly from the app.
-
Metal Ranger
Gaming Apps
Metal Ranger is a 2D shooter with nostalgic influences from science fiction films and games from the 1980s.
-
275 USD DeepChef
AI & Machine Learning
Our challenge was interpreting images of ingredients with varying quality and lighting. By refining our AI model, we succeeded in delivering accurate and creative recipes based on the input.
-
275 USD SnapDecor
AI & Machine Learning
To achieve realistic room transformations, we trained our AI on a vast dataset of furniture and décor. We tackled lighting and perspective issues to deliver seamless virtual designs that inspire our users.
-
275 USD SafeZone
AI & Machine Learning
We integrated cutting-edge object detection and tracking algorithms to accurately identify suspicious activities. The challenge was minimizing false positives, which we achieved through extensive testing and iterative improvements.
-
by TrashTalker
The project.
We faced the challenge of differentiating between similar-looking materials. By training our AI on various waste items, we achieved high accuracy in sorting recyclables, compost, and landfill waste.
-
SafeZone
The project.
We integrated cutting-edge object detection and tracking algorithms to accurately identify suspicious activities. The challenge was minimizing false positives, which we achieved through extensive testing and iterative improvements.
-
by SnapDecor
AI & Machine LearningThe project.
To realistic room transformations, we trained our AI on a vast dataset of furniture and decor. We tackled lighting and perspective issues to deliver seamless virtual designs that inspire our users.
-
by Zakupki.Prom
Cryptocurrency & Blockchain
Problem: the customer needs a 100% data security guarantee Solution: It was agreed with the customer on the replication of the database with a backup on hidden replicas (through an increase in infrastructure operating costs). Result: the customer satisfied this solution and this solution of the problem was implemented.
-
Purolator
Javascript and Typescript
The problem: the customer needs to be served by users in the US, Europe and Australia. SDN CloudFront for file delivery uses server-less architecture that allows you to place the "server" as a lambda function in several data centers. This system was implemented and tested, after testing the necessary directions were introduced that satisfied the customer. Result: the buyer was given this decision that meant his demands regarding this problem.
-
The Domino World
Gaming Apps
The problem is the building system.
Solution: a user interaction interface has been developed with
The details.
Different types of details were developed and agreed with the buyer.
… After that, we started a more detailed processing of each type.
Result: a large number of details and methods of their application
He needed a physical arcade.
Domain behavior model
Solution: a number of physical models have been developed.
I chose the comparison.
Optimization of the physical choice was made.
Tests of the physical model.
Result: a physical domain model that meets the target
Audit of the program.
-
Code of War
Gaming Apps
The problem: the implementation of the network game mode
Solution: Research and testing of different servers and
Synchronization of players in the lobby.
… Choose “Photon Server” because it meets all the requirements
the orderer.
They raised the test game server and conducted tests from different
of regions.
After the test, optimization was made to reduce delays.
Packaging and better synchronization.
Result: Stable servers with small delays even when
Slow speed of the network. The Great Arsenal of Different Weapons
Resolution: The study of modern samples of weapons.
Agreement with the buyer of their graphic and sound design, as well as
physical model of behavior.
Testing the advantages of various weapons in the game process
Implementation of Correction
The result: A large number of unique weapons and its modifications.
