36" x 48" drawing robot
Currently working on the 8.5" x 11" version
Created custom PCB stepper motor drivers
Three Pi Zero Ws with simultaneous cameras and stitching to create a 3D effect
Process and parse Latin sentences with Whitaker's WORDS and some graph analysis
Used both Wikipedia API and a local Wikipedia Dump to traverse pages until the Philosophy article was reached.
Constructed linked data structure in Python to model the shape of linked pages
Used Python to export model to GraphML standard
Used Graphia and Gephi to model Wikipedia as a graph.
Complete Rubik’s Cube solver
Used computer vision to detect colors and positions of each individual facelet
Because of limited resources, devised a method of understanding unseen edges with only one camera
Implemented Kociemba’s Two-Phase solver for efficient solutions
3D printed a frame and motor stems and soldered 6 stepper motors to motor drivers, and Arduino
Wrote a serial communication system between Arduino and Python.
Repurposed computer power supply for motors to ensure electrical stability
Used Python and computer vision to detect faces in an area and point a laser at it
Wired servo motors to a RPi motor hat and external power supply.
Computer vision from MacOS connected to RPi with Python server
Idea based on Michael Reeves’ original YouTube video, but written in completely original code using different libraries, languages, and hardware.
Used Python to create conceptual proof of ray tracing 3D-renderer.
Used input coordinates and linear algebra to detect light-surface collision with triangles.
AI-driven system to detect enemies on a crosshair in CSGO in order to flick away from enemies and make it impossible to kill enemies.
Used PIL to parse a few thousand input images into TensorFlow image neural network.
Made the smallest possible keyboard (6 buttons) by using each mechanical switch as a binary input.
Soldered switches together with diodes for efficient layout system
3D printed keyboard case and keycaps
Used Teensy 3.0 (Arduino) to connect through USB and parse inputs
Used CV to detect if eyes were open through a webcam and would wake the user up if eyes were closed for too long with a 110 dB piezo buzzer.
Piezo buzzer connected to computer vision through a Python server on a RPi relay system.
Created a solver to give all solutions to the daily NYT Spelling Bee game
Used Python to simulate encrypting and decrypting messages in the same style as the Enigma machines used in WWII
Modeled a 8-bit Carry Look-Ahead adder circuit in Minecraft Redstone
Used Python to generate a tree of links based on a given Wikipedia article by parsing JSON from Wikipedia’s API
Used OCR to read type-test words and type them extremely fast (~350 wpm)
Used Processing/j5 to model general physics equations
Used AppleScript to mash the spacebar and type passages
The typing tool gained over 30k views on TikTok
Used AppleScript to spam messages to people and groups in timed intervals.