Hi, I'm Thomas 👋
I help teams ship reliable AI.
From problem to product.
TN

ThinkDraw

Speak your ideas and watch them become mindmaps, diagrams, or sketches in seconds. Turns messy thoughts into clean visuals without the drawing hassle.

Vite
SST
LLM
TTS
tldraw SDK
website

PokeClicker

Built web-based Pokemon game with React and TypeScript. Implemented game state management, MongoDB backend for saving progress, and interactive battle system.

React
TypeScript
MongoDB
Vite
github
website

Bridgable MCP

Built MCP server that gives AI agents instant access to OpenBridge domain knowledge. Integrated RAG system to enable one-line installation for querying OpenBridge documentation and APIs.

MCP
RAG
AI Agents
Python
website

Vervify

Built centralized directory for all NTNU student organizations. Implemented self-service registration system and search functionality to help students discover campus activities.

Next.js
TypeScript
PostgreSQL
github
website

Cogito x Infor

Led team of 9 students building ML recommendation systems for Infor ERP. Developed product recommendation engine and price prediction models using customer purchase history. Built no-code browser test automation tool for cloud ERP, replacing thousands of hours of manual testing.

Python
Machine Learning
AI Agents
scikit-learn
github

Futleie

Built rental marketplace web app with Next.js and Supabase. Implemented authentication, item listing/browsing, and admin dashboard for managing rental inventory.

Next.js
TypeScript
Supabase
Tailwind CSS
github
website

Kollapp

Built desktop app for student shared housing management. Implemented task scheduling, expense splitting with automated settlements, and group chat using Java and JavaFX.

Java
JavaFX
github
Hackathons
  • R

    Recover Norge Hackathon

    1st place
    Built ML ensemble model to predict missing actions in home repair calculations. Trained XGBoost, CatBoost, LightGBM, and custom neural network to automate manual identification process. Delivered three production-ready demos solving real workflow problems.
  • N

    Norwegian AI Championship 2025

    2nd place - Norwegian AI Championship
    Week-long AI competition with three tasks: built RAG system for medical document NLP, trained RL agent for autonomous driving, and implemented computer vision model for tumor segmentation.
  • A

    A* Hackathon 2025 w/ Norges Gruppen & Cogito

    3rd place
    Built real-time YOLOv11 system for self-checkout with automatic product classification, receipt generation, and theft detection. Developed web app with mobile camera integration for live demo.
  • B

    BRAIN Hackathon 2025

    2nd place | Best presentation | Most creative approach
    Built YOLO 11 object detection model for autonomous vessel obstacle detection and collision avoidance. 24-hour Kaggle competition focused on model accuracy, interpretability, and risk assessment.
  • S

    Start Code 2024

    Built automated Raspberry Pi to web integration for emno. Implemented hardware-to-web connection system during 48-hour hackathon.

Work Experience

O

OpenBridge

Aug 2025 - Present
Part-time Software Developer
Working as a part-time software developer while studying, focusing on full-stack development and contributing to innovative software solutions.
R

Rystad Energy

Jun 2025 - Aug 2025
Summer Intern
Summer internship at a leading energy intelligence company, gaining experience in data analysis and software development within the energy sector.
C

Cogito x Infor

2024 - 2025
AI Intern / Project Leader
Lead a team of 4 students in collaboration with Infor, the world's third-largest ERP systems provider. Developed several AI-based solutions, including recommendation systems, price recommendations, and now AI agents to improve business processes. Focusing on machine learning, project management, and practical implementation of AI solutions.
N

Nightlife Industry

2023 - 2025
Security Guard
Responsible for guest safety at various nightlife venues. Developing strong communication and collaboration skills by handling challenging situations at a fast pace.

Education

N

NTNU

2023 - 2026 (ongoing)
Bachelor in Computer Science

Skills

Python
Machine Learning
RAG
Computer Vision
Reinforcement Learning
AI Agents
MCP
Recommender Systems
TypeScript
React
Next.js
Node.js
Java
Git

Motivation

I like building. I like learning. And I like the feeling when a messy workflow becomes a simple product. That's why I spend time on evaluation, reliability, and shipping, not just demos.

Contact

Get in touch

Want to chat? Feel free to reach out via LinkedIn or send me an email at thomanso@stud.ntnu.no.

LinkedIn