A bit more background on how and why I decided to build it.
Building Under the Radar was super fun because I was able to use everything I learned building Product Explorer and http://Opportunities.so.
It's the first time, I’m using a custom-made user interface. Really hoping it will survive this first real world test.
Initially my idea was to do what everyone else is doing when it comes to trendspotting and then use price as the main differentiator.
Primarily, I wanted to have a simple project so that I can learn more about custom user interfaces and data pipelines.
But as soon as I started working on the project, I had hundreds of additional ideas and, well, now the project is no longer that simple.
One idea was that instead of relying on random walks in keyword space, the algorithms are monitoring specific data sources like GitHub.
This not only leads to a more interesting selection of trend signals but also allows me to show them in their proper context alongside relevant metadata.
A common question I got is: "What’s the difference between this and http://Opportunities.so?"
In short: Under the Radar is use-case agnostic whereas Opportunities is specifically for entrepreneurs looking for business ideas.
We're using the Under the Radar data to find interesting signals for our http://Opportunities.so trend brainstorms.
Fun Fact 1: Instead of preparing properly for the launch this morning I thought it would be smart to add a completely new category of trends (crypto).
Needless to say that I also added quite a few bugs.
Fun Fact 2: Instead of a proper database, I'm using Google Sheets.
The main reason is that with Google Sheets it's so easy to manually edit and curate data. I can now easily bring in freelancers without having to build an admin interface.
More on the stack:
The front-end is hosted on http://Render.com. Love the push- to-deploy workflow.
All backend code is Python and runs on a cheap Digital Ocean server.
Everything else, 1.) payments 2.) sales taxes 3.) authentication is handled by Gumroad.