Skip to content
View alextaschuk's full-sized avatar
🧑‍💻
🧑‍💻
5 followers · 10 following

Highlights

  • Pro

Block or report alextaschuk

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
alextaschuk/README.md

Hello, I'm Alex!

I have a degree in computer science from UBC. While I was in school, I gained an interest in backend development and networking.

In my free time, I enjoy:

  • Reading
  • Listening to, discussing, and finding new music
  • Rock climbing

What I'm Currently Working On

Looking for and applying to new grad roles!

Previous Projects

  • Clock-Based Text Detection using Deep Learning Transformers, Token Classifier, OpenAI

    • For a computer engineering class project, I explored the possibility of using machine learning to detect sentences in books that mention an explicit hour and minute of the day. I built two pipelines and compared the results: the first approach used a fine-tuned transformer model (spoiler alert, it wasn't very good). The second approach used a fine-tuned token classifier to tag any temporal text, then GPT-5.4 mini to filter any tagged text that wasn't clock-related (e.g., remove "yesterday" but keep "1:43 P.M.").

  • TCP Congestion Anomaly Detection C, TensorFlow, Keras

    • I implemented TCP from scratch in C (did not use TCP sys calls like socket()), then trained and fine-tuned an LSTM to predict imminent congestion events (triple ACKs/timeouts) before they occur.

  • University Capstone Project Python, SQL, Electron

    • I worked on a team of 6 for a year-long school project to develop a full-stack desktop app for generating resumes and portfolios. Users upload a compressed file (e.g., .zip, .tar) that contains coding projects to the app, and the app will analyze the projects and generate a usable resume and web portfolio with statistics and information about each project.

  • Literary Quote Clock Python, Raspberry PI, E-paper Display

    • I built a clock that tells the time using quotes from books using a Raspberry PI Zero 2WH and an e-paper screen.

  • Selectify (albumselector.com) Python, Nginx, Gunicorn

    • I made a website that selects a random album from your list of saved albums on Spotify for you to listen to. The site was originally deployed locally via a reverse-proxy, Gunicorn, and port forwarding; it was accessible to anyone, regardless of whether they were connected to my home network or not.

    Selectify Demo

Languages

Python, C, Java, JavaScript, R

GitHub Stats

How to Get in Touch

Pinned Loading

  1. Literary-Quote-Clock Literary-Quote-Clock Public

    A clock that displays the time using quotes from books.

    Python 1

  2. TCP-Congestion-Anomaly-Detection TCP-Congestion-Anomaly-Detection Public

    Manually implemented TPC in C and trained an LSTM to detect anomalous network congestion activity for COSC 448.

    C

  3. Clock-Based-Text-Detection Clock-Based-Text-Detection Public

    Explores the temporal tagging performance of a fine-tuned BERT token classifier + GPT-5.4 mini vs. a fine-tuned RoBERTa seq2seq model for CMPE 401

    Jupyter Notebook

  4. selectify selectify Public

    A website that selects a random album from your Spotify account

    HTML 2