/ai
AI & Technology
I bring AI into web applications: agents that do work on the user's behalf, automations, and models I train myself when needed. This page brings that work together.
Areas of work
AI agents for the web
LLM-based agents embedded in web applications that call tools and work with APIs.
LLM integration
Connecting language models to existing products: prompt design, structured output and safe API use.
Automation
Workflow automation with n8n; connecting AI steps to real business processes.
Deep learning & computer vision
MRI image classification, transfer learning and segmentation (U-Net).
AI projects
- 2026NeuroVision AIMRI-based brain tumour classification with five architectures and a Flutter mobile appComputer Vision
- Football AnalystAn analysis app that turns football data into clear results.Data Analysis
- 2026Segmentation (planned)BRISC2025 · U-Net + EfficientNetB0 encoderComputer Vision
Methodology notes
Deterministic data preparation
All experiments share the same seed (SEED=42) and the same cleaned split, so comparisons stay fair.
Measuring leakage
Duplicate cleaning with SHA-256 + pHash, GroupShuffleSplit and leakage assertions in code.
Anatomy-aware augmentation
No horizontal flips on brain MRI; the symmetry assumption can destroy information.
Mistakes are findings too
The EarlyStopping callback carrying state was documented as a methodology note that affected results.
Technologies used
- Languages
- TypeScript · C# · Python · Dart
- Web
- Next.js · React · ASP.NET Core · PostgreSQL
- AI
- LLM APIs · ResNet50 · DenseNet121 · EfficientNetB0 · U-Net
- Automation
- n8n
- Mobile
- Flutter
- Infrastructure
- Vercel · Render · Kaggle (Tesla P100 / T4)