jmedcoff/projects

jmedcoff.com

Winter 2021
Yup, this very website, constructed as an experiment in web dev and hosting from the ground up. This site is 100% handwritten html and css, no javascrapt. Hosted on a raspberry pi. Less than 24 hours after standing up, it's already getting hit with POSTs to nonexistent admin pages. Godspeed.

Fender-style tweed deluxe amp

Winter 2021
Plagued with pandemic lockdown boredom, I got a guitar amp kit from Mojotone. The kit consists of a chassis, cabinet, a couple transformers, and a lot of components (resistors, caps, pots, tubes, etc.) and an instruction booklet. The process of soldering/wiring/building was completed across several weekends, and the result sounds pretty nice. It's a ~15 watt amp, suitable for bedroom/practice use, and is a pretty good pedal platform.

Code Concert

Fall 2018
I returned to GrizzHacks 3 in 2018 a sponsor rather than a participant. I was invited to speak, and choosing to focus on a light topic rather than dense technical material, that typically goes in one ear and out the other at 9pm (at least for me) I gave a talk on a Haskell live coding environment called Tidal. I went over FP, patterns and effects, and gave a demo in the form of about 15 minutes of on-the-spot live music.

Pychov

Fall 2017
At GrizzHacks 2, I worked with back-hopping champion Gocnak on a Discord chatbot written in python. As was true for most of my hackathon work, this project was pretty satirical of typical AI chatbots for two reasons: first by not using AI, and second by involving the internet. The bot scrapes theoretically any website of your choice for textual content, then uses a Markov chain to perform iterative text generation. It's usually incoherent, but very fun.

Fourier-Motzkin

Summer 2017
Serge Kruk, perhaps the greatest educator at Oakland University, advised me on a project within the math department. I wrote code in common lisp (yes!) to take a set of linear inequalities and equations as input, then project the polytope they define onto a subspace using the Fourier-Motzkin method.

Biological Physics

Summer 2017
Before my final year of undergraduate education I had the pleasure of working with Steffan Puwal of the Oakland University Physics Department. The goal of Dr. Puwal's research in general is modeling human cardiac tissue in order to simulate fibrillation, and more strictly, to simulate methods of defibrillation. Over the course of a few months, I wrote some octave code to simulate cardiac action potentials using forward Euler. Then, after some plotting and curve fitting, we were able to measure the result of inputs to the simulation in geometric terms in phase space. The research culminated in a talk I gave at University of Michigan Dearborn.