🗂️ Draft Projects

A collection of code — interview assignments, tutorial follow-alongs, and ideas still taking shape. Nothing here is meant to be production-grade; it's a playground for learning, practicing, and testing things out.

⚠️ Each project lives in its own folder and is meant to be run in its own environment. Check individual folders for extra setup notes where available.

• 🎯 Interview Projects
• 📘 Tutorial-Inspired Projects (roadmap.sh)
• 🥾 Hiking & Location Projects
• 🧪 Other Practice & Draft Projects
• 🛠️ Utilities & Setup
• 🤝 Adding Existing Projects

• LiveFootballWorldCup-java ⚽ — A Java library that manages a live football scoreboard: start a match at 0-0, update scores by team name, get a summary ordered by total score, and finish (remove) a match. Built from an explicit interview-style spec, with assumptions documented in its README.
• LiveFootballWorldCup-kotlin ⚽ — The same scoreboard exercise, implemented in Kotlin with its own set of documented assumptions (e.g. how "matcher" objects are used to update or finish a match).
• doctor-file-processor 🩺 — A Spring Boot service built for a take-home assignment: it periodically scans a directory for JSON documents, validates them, and moves them to valid/ or invalid/ with an audit log. It also exposes an HTTP endpoint for on-demand validation with a custom-built rate limiter (5 requests/minute, no third-party library), tested under concurrent load.

• pybroadcast 📡 — A Python WebSocket broadcast server, packaged as a CLI tool (broadcast-server), that lets clients connect and receive messages broadcast by the server. Based on the roadmap.sh Broadcast Server project.
• ImageProcessingService 🖼️ — A Python microservice demonstrating image handling with Kafka messaging, containerized dependencies, and a logging pipeline (Filebeat → Logstash → OpenSearch). Its README also outlines a planned AWS deployment (API Gateway, Lambda, Aurora, S3). Related to the roadmap.sh Image Processing Service project.
• linux-setup 🐧 — Detailed, hands-on notes from working through the roadmap.sh Linux Server Setup project: creating a sudo user, hardening SSH, configuring UFW, enabling unattended upgrades, setting up Fail2Ban, and inspecting logs with journalctl.

Several related attempts at combining hike-tracking data (GPX files, Strava) with photos.

• HelloWorld 👋 — Despite the name, this is a C# .NET project that parses GPX files and pulls trail data from Strava through a StravaClient — an early experiment that looks like a precursor to GpxClient/HikingClient.
• GpxClient 🧭 — A Python/FastAPI service that parses uploaded GPX files into coordinate points and integrates with the Strava API (OAuth login, token caching via MongoDB, fetching activities).
• HikingClient 🥾 — A C# project whose README documents research into matching hike GPX data with photos: investigating the Strava API (OAuth 2.0, activity endpoints) and the Microsoft Graph API for pulling images from OneDrive.
• HikingClient-py 🥾 — The Python/FastAPI counterpart: includes a Microsoft Graph/OneDrive OAuth login flow (router.py) and a local file-processing entry point (main.py).
• photo-map 📷 — An early-stage FastAPI scaffold (its app title is still "GPX Client") that renders a template via a Wikipedia-client auth provider — a work-in-progress take on combining photos with location data.
• HikingPinpoints 📍 — Currently just a project skeleton (IDE config only, no source code yet).
• hike-map 🗺️ — Also just a project skeleton at this point (IDE config only, no source code yet).

• aws-python ☁️ — A sample Python app deployed to AWS Lambda with DynamoDB, using CDK/SAM for deployment and a GitHub Actions CD pipeline. Its README also covers IAM permissions, VPC networking, and CloudTrail logging.
• cli_project ⌨️ — "MCP Chat," a command-line chat client for the Anthropic API that supports document retrieval (@doc) and slash commands, built around the Model Context Protocol (MCP).
• call_antropic_api 🤖 — A Jupyter notebook experimenting with direct calls to the Anthropic API, with small helper functions for building a message history.
• claude 🧠 — Claude Code configuration: a pr-description skill that writes pull-request descriptions from git log, and an accompanying agent file for more detailed PR instructions.
• fast-api ⚡ — A FastAPI practice project with a basic item/project CRUD API, custom logging and timing middleware, and unit tests.
• oauth-py 🔐 — A FastAPI app demonstrating Google's OAuth 2.0 authorization code flow (via the People API), with OAuth wired directly into the Swagger UI.
• image-sharing 🖼️ — A FastAPI app with user authentication (via fastapi-users) and image upload support (using ImageKit), modeling users and posts.
• postgredatabase 🐘 — A Spring Boot + PostgreSQL/JPA app modeling authors and books, with repository, service, and controller layers plus integration tests.
• database 🗄️ — A minimal Spring Boot skeleton with a JDBC starter dependency — set up, but without business logic added yet.
• database-backup 💾 — A placeholder Python script intended for automating database backups (currently empty).
• py-salesforce-sync 🔄 — A placeholder for a Python service to sync data with Salesforce (currently just a stub).
• mobile 📱 — A small Python script that uses adb to send an SMS from a connected Android device.
• static-content 📄 — A simple static site (sample "fruits" and "vegetables" pages) served via an Nginx config — practice with static content hosting.
• store 🛒 — A Spring Boot + MongoDB app modeling a store: software-engineer listings plus order and payment services (Stripe and PayPal integrations).

• .idea — IntelliJ IDEA project configuration files.
• LICENSE — MIT License covering this repository.

Since each project originally lived in its own local repository, they were merged into this monorepo using local remotes and unrelated history merges:

git remote add football-java ~/Projekti/LiveFootballWorldCup-java/
git fetch

git checkout -b football-java --track football-java/master

git merge football-java --allow-unrelated-histories

Got a project idea you want turned into its own thing? Feel free to fork it out into a standalone repo once it's ready. 🚀