AI Traffic Monitoring
Automated vehicle counting and classification powered by Convolutional Neural Networks, for analysing traffic volume in images, video files and live webcam feeds.
Read more →Researcher · Software architect
Software architect and researcher.
Research and experiments around data, focusing on analysis, representation, and processing to extract insights, preserve accuracy, and reduce computation time and power consumption.
Topics include Convolutional Neural Networks (CNNs), Explainable Artificial Intelligence (XAI), computer networks, storage systems, distributed computing, and data security.
Insight
Performance
Security
Representation
Open Source
Tools and applications
Open-source tools, libraries and applications, from AI and computer vision to traditional software: networking, over-the-air (OTA) updates and edge devices.
AI and computer vision
Medical AI Diagnosis Evaluation
Explainable Artificial Intelligence: Human vs AI Diagnosis Evaluation Application.
Read more →
Handwritten Digit Recognition
Handwritten digit recognizer application powered by a Convolutional Neural Network.
Read more →Networking, edge devices and systems
IoT Data Transmission Suite
A data transmission suite harnessing the widely deployed HTTP/HTTPS infrastructure for a cost-effective IoT ecosystem.
Read more →
otashell
A bash script for Linux systems to perform Over-The-Air updates, enabling remote application and configuration management.
Read more →
tcptunnel
Linux Secure Remote Access Agent - Securely access your local Linux machine from anywhere via SSH reverse tunneling.
Read more →
jsonlib
Simple JSON library in C for embedded devices. Lightweight, minimal, and designed for generating flat JSON objects and arrays with a very small memory footprint.
Read more →
xmlparser
XML parser for embedded platforms with low memory footprint.
Read more →
All Repositories
All 18 repositories: Linux utilities, embedded C libraries, command-line tools and AI applications.
Read more →About
Viki (VN)
Software architect, systems engineer and independent researcher.
Education
Master of Science in Computer Science
Specialisation in Artificial Intelligence, Germany
Background
More than fourteen years in software engineering, including nearly a decade at a major computer networking company building performance-critical networking software on embedded Linux, and the design of telematics and IoT systems connecting around ten thousand devices.
Infrastructure and AI
Alongside AI, the work covers the infrastructure those systems run on: virtual machines, containers, Kubernetes and cloud platforms. Models and data pipelines only create value when they can be deployed, scaled, monitored and secured reliably, so the work follows one thread from Convolutional Neural Networks (CNN) and Explainable Artificial Intelligence (XAI) at the model level down to the networks, storage and distributed systems underneath.
A bachelor’s degree in Computer Science Engineering and master’s research in artificial intelligence provide the foundation for this: drawing insight from data, speeding up processing without losing accuracy, and protecting data in storage and in transit.
Tools and languages
Primarily C, Python and Bash, with supporting work in SQL, PHP and JavaScript. Frequent use of Wireshark, GDB, Docker and standard Linux tooling.
Contact
Questions and collaboration
If you have a challenging problem in data processing, security, architecture or the deployment of critical systems, feel free to write. Research collaborations and solution discussions are welcome. If I can contribute usefully, we can discuss how to proceed.