Introducing Noiseless Signals!


The soft-launch of my new web-app, Noiseless Signals.

Introducing Noiseless Signals!

Better late than never!

I’m proud to share that I have soft-launched my new web-app, Noiseless Signals. As a day-trader, I’m often asked: “Should I buy or sell ‘asset x’?” Well, ask no more. Noiseless Signals takes my daily trading approach and codifies it for the masses. A user can quickly see (and even get alerted!) if asset x is a buy/sell/hold, without all the noise of headlines, rumors, and gossip. Now, the cynics will say that this is simply (self) promotion, and they’re not wrong. But I also want to reflect on the process and highlight what I learned:

Organization matters: Switching back-end API financial providers was super simple as a result.

Refactor: As the app inevitably scales, there are always (fun) ways to optimize efficiency.

Build: I built my own API with rails to protect my API key! This will also serve me well should I ever develop for mobile.

Redis databases: Scheduled alerts for paid users and database management runs with Sidekiq background scheduling on a Redis server. I was self hosting this to save costs but moved it into the same datacenter as my web-server and saw significant performance improvements.

Front-end overhaul: I found a library I liked and stuck with it (tailwindcss).

150,000+ models in a database is a perfect opportunity to creatively optimize.

Fresh air remains a great source of inspiration for challenging problems.

There’s still a lot to build and optimize, but I am proud that I have a functional web-app ready for customers!

This is one of the original iterations:

The app has (thankfully) come a long way from the early days...

To this:


Current (mobile) watchlist view with Lyft alert options expanded

Admittedly, it isn't perfect (yet), hence the soft-launch of the MVP.

Soon, paid users will be able to create, manage, and even import existing portfolios. A one-stop platform for tracking your portfolio and beating the market.

Happy trading!