Thrift Label is a search engine aggregator for second-hand e-commerce listings.
Building off the idea of Futureshop, Thrift Label is a search aggregator that combines different marketplaces, but just the second-hand listings.
By combining the search results for many big and small e-commerce sites into a single search experience, Thrift Label makes it easier to find great second-hand deals on just what you are looking for.
Mega-marketplaces such as eBay and Amazon ensure that even long-tail searches will often return relevant results.
A data explorer and visualizer built off the NYC Open Data portal.
NYC Open Data is an expansive portal of over 3000 datasets, files, maps, and more. I built a portal to quickly search, sort, and skim the datasets, and open them up for basic exploration.
This project is a little bit unique in that my goal was not do dive deeply into a specific dataset, rather it was to cleanly and intelligently display any of the datasets.
To start, I had to scrape the metadata from all 3000+ datasets. After first using Playwright to render the javascript, I found a method that didn’t require javascript, and instead parse the text with regular expressions.
I chose Streamlit as the visualization package, largely because it looked cool and feature-rich.
For the viewer page, I wanted the page to be able to detect columns that were latitude and longitude, since the datasets do not have consistent column names. With those columns detected, I could map any datasets that contained a lat+long.
There are many features that I’d still like to add, especially around the handling of other types of datasets, such as “map”, which is sometimes geojson and sometimes csv, and thus a bit more complex to process.
Project requirements:
Scrape meta-data from datasets
View, display, and sort meta-data
Viewer to load and display dataset data
Automatic recogniction of lat/long columns, conversion to numeric formats, and display on map.
Brode Vitamins reinvented the sugary sports drink as an easy-on-the-go electrolyte tablet.
As a college athlete, I had a honed sense of what it meant to be hydrated. Sports drinks had too much sugar and oral-rehydration-solutions (Pedialyte) were expensive–and quite gross tasting.
Starting as a DIY project to reverse-engineer Pedialyte, I developed Brodes as a way to turn ORS drinks into a convenient portable tablet format. Brode Electrolyte Vitamins are now over 10 years old, and are doing better than ever.
Brode is set up as an automated e-commerce operation, with a D2C Shopify site and Amazon listing. Fulfillment is handled by a 3PL and production by a trusted manufacturing partner.
Futureshop is a search marketplace for clothing made more sustainably, with lower environmental footprint.
I’ve always been a very “conscious consumer”, but I also am acutely aware of the extra effort it takes. I don’t mind the effort, of course, but I know lots of people who would love to shop more “sustainable” if it were just at least almost as convenient as otherwise.
The thesis behind Futureshop is that if customers are given a broad search experience of pre-qualified products, they will use it.
Initial prototyping was built on the Bubble no-code system, before switching to Django for extensibility, control, and scale.
The Django front-end accesses a MongoDB database using Atlas Search. The database is populated by a custom-built Scrapy scraping platform.
Magnet Jar was a space-saving ultra-strong magnetic spice jar.
I created Magnet Jar after being unsatisfied with available spice solutions.
My primary design goals were space efficiency and speed of browsing to find the spice that you needed. I settled on a design with a side-mounted channel magnet that could be stuck to most fridges and any magnetic panel.
This design would engage underutilized vertical space, and prominently display spices to avoid the “out of sight, out of mind” problem where spices get stuck in the back of a cabinet or drawer.
Existing magnetic containers had magnets on the lid, which I did not like. I likewise did not find a spice rack or shelving system that I liked, which either took up counter space or did not firmly keep jars organized.
After selling some thousands of Magnet Jars on Amazon, and earning a strong 45% returning-buyer rate, I decided to discontinue Magnet Jars in May 2020. (I assembled every jar and package by hand.) Profits were not high enough to justify the effort, and I did not see a path with enough certainty to justify expanded investment.
Although Magnet Jar remains maybe my favorite product, and I have thirty of them on my fridge right now.
Rustic candles to repurpose waste lipids, and harken back to a bygone era
Years ago, households made something called a “prairie candle” using tallow rendered from animals. Candles are a wick surrounded by a fat/lipid, and households would use what they had, which was often tallow or even lard.
actual photo of some of my family members, who probably made prairie candles.
One Thanksgiving season, a family member killed a deer. After saying a short prayer and thanking the deer for their sacrifice, I asked if I could take and render the tallow. I read that you could make grease, polish, candles, and other things.
I decided to make candles. The first batch made 5 candles, set in mason jars. They worked! The best story from the first batch is that my cousin’s dog ate one.
I found a local game processing ‘facility’ (every locality has a guy), got some new inputs, and made a second larger batch of candles, this time with scents and a variety of wicks. I wanted to test them out.
It turns out that the candles do not perform as well as the candles that we are used to–candles made from paraffin, beeswax, or even soy.
The first brand vision was to create a high-end “rustic outdoors” brand and sell the candles at a high price.
I still like the idea. I like taking any “waste” product and turning it into something useful.
My future vision is to do a different product from the same input, such as cute fire-starters, and have them be made by local non-profit kids groups, like boy and girl scouts, as fundraisers. (Candles would still be an option, too.)
The vision would be to setup the product as a “kit” that local groups could follow the directions to make and sell, potentially under an umbrella brand.
Spacephones were a prototype isolation heaphone especially for a more enjoyable spoken-word audio experience in loud settings.
Have you ever turned up your headphone volume to hear over loud noise, and then later, when you are in a quieter place, you turn on your headphones again and volume feels like it is set at jet-engine levels?
That’s how loud you were listening to the volume before, and that’s not healthy. The sound levels in the NYC subway system can breach 100 decibels. And when you turn up your headphones to hear over it, you’re essentially doubling the sound.
I wanted headphones that would block out the outside noise so that I could maintain a modest volume level. There are some specialty headphones for drummers, then there are “active noise cancellation” powered headphones.
The prototypes were made by taking apart professional-grade hearing protectors, off-the-shelf headphone drivers, and then recombining them together into one product.
They worked great (I still use them), but I was concerned that the market for such high-performance sound-isolation headphones would be limited. The product drawbacks were the large size and the tight fit.
spacephones prototype branding exercise
Spacephones are a fun product that I use to this day, and were an opportunity to practice some soldering and prototyping, while creating another product-market outline.
Meal Mix was blend of whole grain and protein powders, flavored with cocoa and cinnamon, to be blended into shakes.
Before the modern “Soylent” drink, I was experimenting with deconstructed nutrition, but in a whole-foods model. The experiment was to take the ingredients of nutritious whole-grain foods and blend them in a manner such that when you mixed them into a shake, they would taste…good enough.
I experimented with the blend, and put together a complete nutritional analysis, including micronutrients. In case you are wondering, the star of the blend was oat flour, which has both a strong nutrition profile, and a pleasant semi-sweet flavor.
spreadsheet calculating different nutritional profiles for different blend ratios
I tried both 3-pound bagged product packaging, and pre-mixed bottles.
The shelf-life of the bottles was clearly not long enough. After selling some dozens of the 3-pound bags to “early adopters”, I decided that the flavor really just wasn’t where it needed to be.
Personally, I repurposed the remainder of the Meal Mix and made weekly batches of muffins, and at the office I became the muffin-guy.
After seeing where the modern incarnation of “meal replacement shakes” has gone, I’m happy with setting this project aside.
