My homepage

Movies antirecommender

Results:

Select a movie to learn what movie you won't like...maybe. They may still be good movies, so why not try? Start writing and select a movie from the list.
Important: year in the brackets is only a suggestion and before you hit search you need to set year to the correct value.

This website

FART-based taste predictor

Movies antirecommender

A cloud-native simple movie recommendation system that helps users discover films outside their usual preferences. The system uses machine learning techniques (clustering) to analyze movie genres and ratings, providing diverse recommendations that break away from typical viewing patterns.

Key Achievements:

Implemented a fully automated CI/CD pipeline using GitHub Actions
Created unit tests with pytest for both API and core functionality which are run on every push. Deployment launches only if all tests pass
Designed and containerized the application using Docker, creating separate containers for data preprocessing pipeline and the recommendation service
Deployed the application to Google Cloud Run for communication with central website service
Established secure deployment practices using GitHub Secrets for sensitive credentials management
Built final application using FastAPI endpoints
Built local development environment using Docker Compose with user interface in flask and minimal frontend
Implemented comprehensive logging and error handling throughout the application

Tech Stack:

Google Cloud Platform • Cloud Run • GitHub Actions • Docker • Docker Compose • Python • FastAPI • CI/CD • Scikit-learn • Pandas • NumPy • Flask • Principal Component Analysis • K-Means Clustering • GitHub Secrets • Pytest • Git

This website

A cloud-native portfolio website showcasing my projects through a microservices architecture. Each project runs as an independent service in Google Cloud Run scaling to zero, to keeps them cost-effective and scalable. GCP infrastructure was first created manually and is currently fully migrated into Terraform management.

Key Achievements:

Designed and implemented a secure microservices architecture on Google Cloud Platform
Migrated the infrastructure to be managed by Terraform
Containerized the webpage using Docker
Set up automated CI/CD pipelines for terraform sync and GCP deployment using GitHub Actions
Implemented container-based deployment with Docker
Established secure service-to-service authentication using GitHub Secrets and GCP service accounts
Leveraged AI-assisted development using Large Language Models to accelerate frontend implementation and improve code quality

Tech Stack:

Google Cloud Platform • Cloud Run • Terraform • Docker • GitHub Actions • CI/CD • Microservices • Flask • Container Registry • Service Authentication • GitHub Secrets • Python • LLM-Assisted Development

Dominik Suwala

Movies antirecommender

Results:

This website

FART-based taste predictor

Movies antirecommender

Key Achievements:

Tech Stack:

This website

Key Achievements:

Tech Stack:

FART-based taste predictor