Data Scientist and AI Engineer.
I'm Yovan Alvianto, an AI Engineer and Data Scientist from Indonesia. I design and develop AI systems, machine learning models, and data-driven applications that solve real-world business problems and enable better decision-making.
Featured Projects
Explore real-world projects in data science, software, IoT, computer vision and AI.
VialVision
An edge-AI computer vision system using YOLO on Raspberry Pi to detect and classify vial conditions.
Genome Visualization
Built a reproducible Python pipeline to generate publication-quality comparative genome figures.
Volto - Energy Monitoring
Implemented a real-time electricity usage monitoring dashboard with forecasting and anomaly detection.
Used Car Price Prediction
Deployed a machine learning model via REST API to predict car prices based on features.
E-Commerce Price Tracker
Automated product price scraping from online stores for real-time competitive analysis.
Prodigy - Production System
Developed a full-scale production monitoring system to improve decision-making on the shop floor.
SpareSync - Inventory Management
Developed a web-based inventory system for small businesses to manage stock and orders efficiently.
Trashr - Smart Trash Bin
Built an IoT-based waste management system with fill-level sensors and cloud data logging.
What I Can Help You Build
Turn raw data into clean datasets, dashboards, and decision-ready insights.
Build machine learning systems that run in production via APIs, apps, or edge devices.
Detection, classification, and measurement from images or video for real-world use cases.
Digitize workflows with dashboards and operational systems that improve visibility and efficiency.
Practical consulting, architecture reviews, and hands-on training for teams and individuals.
Recent Writings
Insights and case studies from our projects, research, and client work.
Training Data Distribution and Generalization Performance in Real-World Computer Vision Applications
Understanding how training conditions, deployment environments, and distribution shift influence computer vision performance.
General-Purpose vs. Domain-Specific Models: Understanding the Trade-offs
Their versatility has led many to believe that a single foundation model can solve virtually any AI problem.
Understanding the English Advantage in Large Language Models
Multilingual does not imply equal performance across all languages. These models consistently demonstrate superior performance in English compared to many other languages.