Thesis project in AI, decision support and unmanned stores

My thesis explores how an IT-based decision support system can be designed to identify and assess risk situations in unmanned stores.

The work focuses on combining artificial intelligence, multimodal data and analysis models to detect unusual or suspicious behavior in an environment without traditional staff supervision.

A central part of the project is studying how video, access data and transaction data can be combined to create a more complete understanding of what happens inside the store.

The thesis also considers information security, privacy, access control, secure storage and relevant legal and ethical requirements related to camera surveillance and GDPR.

Cybersecurity: network traffic and threat indicator analysis

I have worked with recorded network traffic to understand how suspicious activity can be identified in practice. By reviewing PCAP files in Wireshark, I followed communication flows, interpreted protocols, detected unusual behavior and investigated signs of possible data exfiltration.

The work combined manual traffic analysis with supporting tools for enriching suspicious artefacts, such as checking indicators and decoding data. It gave me a clearer understanding of how technical traces can be connected to security events and investigated step by step.

Dual boot with GRUB and Apache2 web server

A growing interest in Linux led me to install Ubuntu on my computer in a dual boot setup using GRUB.

As part of the project, I set up my own Apache2 web server and built a personal portfolio site. This gave me practical experience with Linux, web server configuration, file management, networking and publishing a website.

To make the solution more secure, I used a Cloudflare Tunnel instead of exposing the server directly to the internet. Later, I moved the site to Cloudflare Pages and connected it to GitHub for a more modern CI/CD-based publishing workflow.

Home Assistant integrated home

I installed Home Assistant on a Raspberry Pi and used an iPad as a fixed control panel at home.

The setup included smart lighting, humidity monitoring, sensors, traffic camera integrations through APIs and a shared digital overview for schedules and public transport.

The project gave me a stronger understanding of how smart home automation can solve practical everyday problems.

Gym application in C# .NET

Together with two classmates, I built a C# .NET application with separate interfaces for staff and members.

The staff interface handled classes, equipment and lending of training items. The member interface allowed users to book sessions, view history, points and calories burned, and search for other members.

We used the MVVM structure with separate layers for presentation, business logic, entities and data access, including Entity Framework Code First, LINQ, Unit of Work and Generic Repository patterns.

Semi-finalist in CyberChallenge 2025

I participated in CyberChallenge 2025 and reached the semi-final together with my group.

The competition involved challenges in cybersecurity, web security and national security, and gave me the opportunity to test my skills and learn from experienced professionals in the field.