Nov 19, 2021
Interns or Entrepreneurs? How AI Camp Redefined the Tech Internship
We thought we signed up for a conventional tech internship. We were wrong.
(This article is written by Jimmy Zhang, Jeremy Kintana, Ray Cai, and Zack H)
The Origin of Cohort 0
Five high schoolers with minimal software engineering experience scattered across three different time zones. A ten week deadline. And a novel software product waiting to be built.
We (Jeremy, Ray, Zack, Jeremi, and Jimmy), the five members of “Cohort 0,” constitute the inaugural cohort of AI Camp’s Talent Incubator internship program, a ten-week product sprint designed for ambitious high schoolers like us.
However, whatever you think you might know about a “normal” tech internship, we implore you to throw all that out the window before you read this article. This way, you, just like us, will have no idea what to expect as we plunge headfirst into the intrigue and thrill of these next ten weeks.
Ideation
Our journey began on July 27th, during the last half of summer vacation. At our first meeting, we said our hellos, assigned roles, and set expectations. We agreed to meet twice a week on Mondays and Wednesdays to share updates and delegate tasks. Many of us also opted to attend AI Camp’s official tech team meetings that took place on Friday afternoons. In time, we realized that we were about to build a project that actually mattered — one that could have actual, widespread influence.
Ultimately, our goal was to create a minimum viable product (MVP) by the end of the ten weeks. We were presented with two project prompts: an auto-updating database of technology internships for students OR an auto-updating database of scholarships for students. After weighing the merits of both options and doing our research, we decided to go for scholarships: there was simply more material to work with, and it was a much higher trending search on Google. Notably, there were problems with current sites that could be improved: sites seemed difficult to navigate and required a ton of fields to be filled out.
Market Research and Product Development
Set on the idea of creating an auto-updating scholarships database, we commenced the next phase of our project, to gather as much data as possible. Making data based decisions is one of the key pillars of AI Camp, after all. This entailed reviewing what the competition did well and where they fell short. In the following two weeks, we scoured scholarship subreddits, dissected current scholarship sites, and sent out our own surveys to gauge consumer interest.
However, with a myriad of established scholarship websites already out there — Scholarships.com, Niche.com, and Bold.org, just to name a few — we knew that our product needed to do something novel and unprecedented, or else we would have no chance of standing out.
But what novelty could our product possibly have? Cooler features? Cleaner UI? A larger scholarships database?
Without a clear idea of even what would even make our product special, we haphazardly jumped into product development: Zack and Ray began creating a design mockup of what our scholarships website might look like, while Jeremi began outlining the project backend.
Back to the Drawing Board
Just as we thought we were gaining momentum, everything came crashing down. After we pitched our idea to the AI Camp tech team during our third week, they did not hesitate to share their reservations and concerns about a scholarships database…
“Is there really a need for more scholarship websites?”
“What will your product do differently?”
“Hasn’t this problem already been solved?”
The tech team hurled question after question at us, to which we nodded our heads in begrudged silence.
Slightly discouraged, we reconvened at the start of the fourth week, with the same question continuing to haunt us: “What would make our product different?”
Well, maybe nothing at all. All the time we spent on the hasty rudiments of product development should instead have been spent refining one concrete, sure-fire product idea. We were too focused on getting our project off the ground, when we should have been asking ourselves if our product could even fly. After four weeks of walking in circles, we finally conceded that we had to change routes.
Jeremi, our backend engineer, offered one suggestion. Noticing that nearly all of our team members were preparing to apply to college, Jeremi highlighted a common issue among us: college emails. College Board’s Student Search Service is what connects students with college opportunities and information, some of which is extremely valuable. However, this too often comes in the form of hundreds of emails flooding students’ inboxes every week.
Without a college email filter (or a comparable solution) available in the market, we decided that we could help high school students, just like ourselves, manage their inboxes by creating our own “college email despamifier.”
Full Sprint to the Finish Line
Finally, we had an idea that we could stick to, but with just over five weeks to go and not a single line of code yet written, we knew we needed to get going fast.
However, just when things seemed bad already, Jeremi had to take a leave of absence. Now we were down to just four members without a backend engineer, and school was starting to pick up. We just couldn’t catch a break!
After a hasty reassignment of roles and a much-needed extension of our internship to thirteen weeks, we started to gain traction. First, the newly formed backend team collected data from a college search website using Beautiful Soup web scrapers, while the frontend team concurrently worked on the Gmail Add-On display. Next, we compiled the scraped college data into a SQLite file and sorted out the filtering criteria and logic for our spam filter. Subsequently, Ray, our new primary backend engineer, took up the cumbersome task of building the entire Django REST API from scratch. Before any of us could rest, we then faced the challenge of connecting the backend, the frontend, and the database. With only two weeks left, we all put in extra hours to integrate the app with the brain of our product, uniting the REST API, the SQLite database, and the add-on display. By the twelfth week, everything was miraculously finished: the kinks and bugs were all worked out, and our product was, at long last, deployed.
Final Product and Presentation
So now, after thirteen weeks, an impromptu restart, and a boatload of constructive criticism, where have we, Cohort 0, ended up? Well, for starters, we finished building a working minimum viable product, just as we initially intended: a Gmail spam filter that is able to sort college emails into “Match” and “Non-Match” categories based on user preferences.
After our Week 13 demo to the rest of the AI Camp team, many of whom were high schoolers themselves, our product received unequivocal praise.
“This is a product we would actually use” echoed across Zoom chats and Discord channels.
Finally, we had done it. An internship that turned into a startup simulator. An idea that wasn’t meant to be that turned into one that was. There was nothing more fulfilling than seeing that despite the countless setbacks and obstacles, we achieved our goal: not just to build a software product, but to build one that people can and would actually use.
All things considered, there is so much that we learned over the last ten (well, thirteen) weeks. Just looking at the technical skills themselves, the list is extensive: Django, SQLite, Postman, web scraping, Git, Github, Google Apps Script, website hosting, Figma, Notion, and many others.
But beyond the tangible technical skills, these last thirteen weeks have taught us perhaps an even more instructive lesson about what it means to be a software engineer.
“AI Camp treated us not as interns, but as leaders and entrepreneurs who would take ownership of their own product.”
AI Camp has always prided itself on being a company created by students and for students. We, Cohort 0, can proudly say that we are a testament to AI Camp’s motto.
So, thank you to Alex Zhou, Alex Duffy, and the rest of the AI Camp tech team for all of their support and feedback throughout this turbulent, but worthwhile journey. Our cohort represented the first of many more internship cohorts of promising young talent powered by AI Camp. And well, for you at home reading this, maybe your product idea will be the next big thing in tech. We, Cohort 0, will share with you the one key takeaway from our internship: whether it takes thirteen weeks or ten years, one idea or a hundred, a full development team or just yourself, go for it, and we at AI Camp will be eagerly waiting here to celebrate your success.
Links to AI Camp, LinkedIns, and Product Landing Page
If you are someone who’s motivated, creative, and eager to learn about machine learning and entrepreneurship among (many) other things, CLICK HERE to submit an application to AI Camp’s Scholarship and you could receive a partial or even full scholarship for Summer Camp 2022!
Also, below are the links to the AI Camp website, our LinkedIns, and the College Despamifier landing page.
