WGRR to Spotify Playlist Sync + AI Insights
This project syncs the WGRR 103.5 radio station's recently played songs to a curated Spotify playlist. It scrapes real-time track data using Playwright or Axios, stores it in a PostgreSQL database, and updates a playlist via the Spotify Web API. It also exposes a web interface for exploring recent plays and music trends.
AI Insights: Users can ask natural language questions (e.g., “What artists were played most last week?”), and the system uses OpenAI's GPT-3.5 to translate queries into SQL and return answers based on the database — a lightweight, scalable analytics layer.
Technologies used: Python, Playwright, PostgreSQL, Spotify API, Flask, Jinja2, Chart.js, OpenAI API.
Targeted Job Monitor
This tool monitors new job postings from Databricks and 53rd (or any other company I'm interested in, in real time. It uses Playwright or Axios to scrape job listings, caches previous results in PostgreSQL, and sends real-time alerts via email whenever a new relevant position is found. Filters are applied to ensure only roles that match my skill set and location preferences are surfaced.
Technologies used: Python, Playwright, Axios PostgreSQL, Regex, Email SMTP, Cron scheduling, HTML parsing.
Clickstream Tracking with Kafka + ClickHouse
I built a real-time clickstream pipeline that captures web traffic using mitmproxy, streams it into Apache Kafka, and stores it in ClickHouse—a high-performance columnar database—via a custom Kafka consumer.
The entire stack runs in Docker containers, and I visualized the data with Grafana dashboards to track trends like most-visited domains. This project showcases my ability to work across the modern data stack—from ingestion and streaming to storage and real-time visualization.
Technologies used: mitmproxy, Apache Kafka, ClickHouse, Docker, Grafana, Python.
