Skip to content
View manumanuk's full-sized avatar

Highlights

  • Pro

Block or report manumanuk

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
manumanuk/README.md

👋 Hi, I'm Aryan.

I'm a Mechatronics Engineering student at the University of Waterloo with an interest in machine learning and embedded systems. Here's a list of some things I'm currently working on and projects that I'm proud of:

  • UW Orbital Firmware: At UW Orbital, I work as a Team Lead and Communications Subteam Lead. On the comms team, we're developing a UHF radio link, which includes building a SPI driver and network stack (AX.25, FEC, AES128) for the CC1120 transceiver IC onboard.
  • NLP Next Word Predictor: An NLP auto-complete tool that predicts the next word you'll type. It uses Google's Universal Sentence Encoder to convert n-grams of size 3 into 512-dimensional vector encodings that are fed into a trained TensorFlow neural network, which then outputs a prediction for the next word from its dictionary.
  • CS50 Introduction to Artificial Intelligence: Over the quarantine lockdown, I spent my time learning AI fundamentals at Harvard University's CS50 course on EdX. Check out the projects I completed for the course, from a crossword puzzle solver to an image classifier.
  • Computer Vision @ UWARG: As a CV team lead at the Waterloo Aerial Robotics Group, I helped create a vision system for an unmanned aerial vehicle that can detect and geolocate targets on ground in real-time. I created a Chrome Extension to simplify data collection using Google Images, trained a YOLO model, and implemented multithreading into the pipeline.
  • Sendable: Sendable is a web app designed to make privacy more accessible to everyone by simplifying PIPEDA requests for consumers. It takes in email data, determines the services you're subscribed to, scrapes the web for those services' email contacts, and then allows you to send a message asking for your data to be deleted. The site was created with Angular, Firebase Tools (Authentication, Cloud Functions, Realtime DB) and Google Drive API.

Feel free to connect with me on LinkedIn!

Pinned Loading

  1. UWOrbital/CC1120Driver UWOrbital/CC1120Driver Public

    Comms CC1120 Driver

    C 2

  2. WordPredictorPLT WordPredictorPLT Public

    Next word predictor created with Flask, React, NLTK, TensorFlow, and more

    Python

  3. CS50-AI-Course CS50-AI-Course Public

    Course projects from Harvard University's EdX course, Introduction to Artificial Intelligence

    Python

  4. UWARG/computer-vision-python UWARG/computer-vision-python Public

    Autonomy on airside system

    Python 17 38

  5. SendableCompSci SendableCompSci Public

    Angular-based PIPEDA Request Automator

    TypeScript

  6. CanvasAppPLT CanvasAppPLT Public

    Canvas drawing app created with Angular

    CSS 1