Hackdays
Fun fact: The term ‘hack day’ predates computers by a couple hundred years. For a lumberjack, every day is a hack day :)
Hackdays can tell you a lot about a company. Whether a company runs a hackday or not is a pretty good tell about how they see their developers. Another good indicator? How they run their hackdays, which can give you an idea of what you might expect being a part of the team.
In my time at Nudge, I’ve been lucky to be a part of three hackdays so far (with my fourth one coming up!), so here’s a glimpse into what you might expect at a Nudge ™ hackday!
Hackday 1: Re-Imagining Registration
The word “registration” can evoke quite a familiar image: a form, an eagerly awaiting textbox for your username, email, and password. Perhaps an annoying “password strength” indicator that never seems to be pleased. Your verification email that never seems to arrive, or for some reason ends up in your spam folder. As is increasingly the case on our mobile devices, but not so much for our computers, what if you could just show your face and get logged in on facial recognition alone? What if you could experiment with a “passwordless” login system? What if you knew very little about setting up an entire login system, and wanted to just experiment with using “Sign in with Google”? Now, imagine having free time to roll with any of these ideas, which is exactly what happened at Nudge’s first hackday of 2019. One of our developers found, while tinkering with a possible facial recognition solution, that her face looked 54% like her. Another developer blew our minds with a proof-of-concept implementation of a passwordless login solution using temporary ‘access tokens’. And yours truly, ended up taking the day implementing something that may sound simple, but was a full day of learning about a somewhat unfamiliar tech stack.
Hackday 2: AI and Machine Learning
We’ve recently been ramping up our knowledge & looking into where AI and Machine Learning can fit into how we work. These also happen to be common buzzwords in the tech industry, so it can be quite enlightening to take a deeper dive into these topics and see how “the sausage is made.” We jumped into a few different zones here. We experimented with sentiment analysis, and got a barebones analyzer running — ignoring the fact that this analyzer determined that every phrase was “negative” (I mean, we’ve all had those days). A member of our data team explored using a Python library to generate word clouds out of some of our customer-facing reports. Another member of our data team gave us an in-depth view into aspects of machine learning — including feature engineering, what a “model” looks like, and how you might tune or test a model. Possibly the biggest, and one of our favourite outcomes was one of our developers working on a proof-of-concept to show what a global search feature might look like in our CMS. We love this story in particular because we went from a small hackday project that recently got released to all our CMS users as a full-blown feature over the course of a few months. We’ve since received some of the most heartwarming feedback you could imagine.
Hackday 3: Optimization
Have you ever had a workday where you had one too many tasks competing for your attention, and in trying to address all of them at the same time, you end up being less efficient at each one? That’s how some of our API endpoints felt too, so we decided to run an optimization hackday. The great thing about tools like NewRelic is they can give really granular information on what part of your app is a) the most used, and b) the slowest. Sometimes these two coincide, sometimes they don’t. We used this as a guiding light to try to optimize different parts of our codebase. There was no requirement to produce something deployable, but rather in the words of one of our senior developers all that matters is that you “make something, break something, or learn something.” Over the course of this hackday we actually ended up with some excellent speed improvements across our app, and having a whole day to dedicate to something like this paid dividends in both personal learning as well as product value.
There is value in the ideas that can arise from a hackday, even if they don’t “lead anywhere” specifically. Freedom, with a few constraints, can be a great catalyst for creativity, and letting people see where their ideas take them has lots of benefits, including but not limited to: boosting their intrinsic motivation, expressing themselves to those around them in a way that regular daily conversation or interaction just can’t convey, and possibly giving you a perspective that may have previously been undiscovered to you.
Interested in joining Nudge? We’re hiring!
Author: Shahan Ahmed
Server-side Developer at Nudge
