For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
S
Serhiy P.

Serhiy P.

AI Trainer & Data Annotator – Fintech, Software Engineering & Computer Vision

Poland flagKraków, Poland

Key Skills

Software

RoboflowRoboflow

Top Subject Matter

Finance – Banking, Payments & Fintech
Software & Technology – Software Engineering & Mobile Apps
Computer Vision – Document & Barcode Detection

Top Data Types

ImageImage
Computer Code ProgrammingComputer Code Programming

Top Task Types

Bounding BoxBounding Box
PolygonPolygon
ClassificationClassification
Object DetectionObject Detection
Computer Programming/CodingComputer Programming/Coding

Freelancer Overview

I am a Senior Android and backend engineer with 10+ years of experience building secure and scalable fintech applications using Kotlin, Java, Android SDK, and Spring Boot. My work has included designing KYC onboarding flows, integrating payment APIs, and implementing strong security with encryption and authentication, which requires careful testing, validation of edge cases, and clear documentation. I am now interested in AI training and data quality work, where I can use my engineering background to evaluate and improve model outputs. I am comfortable reviewing and ranking code solutions, checking step‑by‑step reasoning for technical tasks, designing realistic test cases for mobile and backend systems, and writing precise, well‑structured feedback in English to help models produce reliable, high‑quality answers.

Labeling Experience

Monopoly Card Barcode Region Detection Dataset (YOLOv8‑OBB)

ImageImageBounding BoxBounding Box

I created a custom image dataset to train a YOLOv8‑OBB model to automatically detect the barcode region on Monopoly game cards in live camera frames. The project involved designing the data collection setup, capturing images under different angles, lighting conditions, and backgrounds, and manually annotating the barcode area on each card with consistent bounding boxes suitable for oriented object detection. I defined clear labeling guidelines, file naming conventions, and train/validation splits to keep the dataset clean and reproducible. The resulting dataset was used to train and evaluate the model as part of a Monopoly companion app prototype that can quickly locate and read card barcodes during gameplay.

2026 - 2026

Education

U

Uzhgorod National University

Master, Physics

Master
1992 - 1998

Work History

U

Ukrainian Defense Sector

Senior Android Developer

Kyiv
2024 - 2025
D

Dyvotech

Senior Android Developer. Team Lead

Kyiv
2018 - 2023