More Projects:
My Favorite Woodworking Techniques
Projection Microscope (Free Download!)
Making Adventure Simulators for Friends
Making AI Tools for Friends
Generative Jewelry
Generative Lighting
3D Scanned Artifacts
Generative Planters
Analytics and Machine Learning:
I’ve also spent a few years working with analytics, handling data for companies like Olipop, Dr. Bronners, Clif Bar, True Classic Tees, and more. I studied Machine Learning at Flatiron in 2022 specializing in text and language analysis. I’ve done a number of projects applying LLMs to do things like organize internet browsing history, characterize the virality of information, analyze research paper abstracts, and analyze market conditions to generate large numbers of business or product concepts.
Click these to read about some of these more code-centric projects:
Analyzing machine learning communities on Twitter with three different natural-language processing strategies, and then using six different machine learning algorithms to see how much you can predict which tweets and topics are most likely to go viral
Building a machine vision system using deep learning techniques to detect pneumonia in children’s x-rays
Looking at large amounts of real-estate data in Seattle to identify key predictors of home value, and then use a linear-regression algorithm to identify homes that may be undervalued in the market & plot them on a map
Experiments in making AI tools that friends and family actually use
Visualize your web-browsing history as a network of connected topics and websites. An AI pipeline interprets internet browsing sessions and feeds these descriptions into a graph-based notetaking software called Obsidian to map out your habits and interests.
An experiment using an ensemble AI to analyze changing market conditions, come up with product definitions, and generate product or business plans
An experiment in extracting features from whitepaper abstracts to see how much explanatory power they have to predict research impact
Using machine learning systems to interpret data from the Taiwan stock exchange to identify or predict companies which are at risk of bankruptcy.
A pipeline for identifying and visualizing the important semantic dimensions of text, like seeing how themes in books evolve, or understanding how your tweets and writing changes over time
What is generative design?
Generative design is the use of mathematic functions to create complex shapes. Generative design has been used for years in computer graphics for CGI in movies and games, sometimes it is used in architecture. It’s almost never used for everyday things, and that’s what I’d like to change.
Generative design can be used to create recursion, textures, fractals, organic forms, and a limitless number of complex geometries. It’s used all around us in nature in the way plants grow or snowflakes form. It’s a broad category that could be expected to become more common over time as computation continues to become more available.
Generative design also allows continuous variation. By structuring designs around variables that constantly change, we can make every item a unique shape.
These techniques can be used in addition to traditional design methods, meaning a custom piece can leverage generative design to incorporate elements of nature or intricate details into something which is hand-designed.
What is 3D printing?
3D printing is a flexible way to make things by precisely adding material. I use Fused Deposition Modeling (FDM) and Stereo-Lithography (SLA) for 3D printed objects or contract pieces. These technologies weld polymers using heat or light to to build up complex shapes in layers. There are other 3D printing technologies, but right now these are the most mature and cost efficient approaches for creating designs around human scales.
You can think of 3D printing as similar to other crafts in that it’s a nuanced discipline with a range of tools and techniques that can take years to master - but it’s not as daunting as it sounds, and basic 3D printers can now be purchased for around 200$.
I use PLA Biopolymer for all of our FDM items, which is a degradable plastic made from fermented plant starches.
What is 3D Scanning?
3D Scanning is the process of converting physical items into spatial data. There are a range of 3D scanning technologies using lasers, projectors, and cameras, each with benefits and drawbacks and startup costs ranging from tens of thousands of dollars to free.
3D scanning is an exciting and diverse area with broad applications in arts, sciences, and engineering.
Replicating advanced scientific imaging systems for under 100$