Course Page Screenshot

pitch

Inspiration

The reason I developed Specters was because, I realized there has to be a better way to learn things. When I was learning Tailwind, I was studying the documentation, forum posts, and consulting AI. I was thinking, "Why isn't there one place I can learn all these things?", and then, Specters was born.

What it does

Specters generates courses with Gemini 2.0-Flash on user specified prompts/topics. Each course consists of 3-8 Chapters (Parts) and each Chapter has 3-5 components. To move on from one part to another, the user must complete a multiple choice quiz on the current part and score 50% or higher.

How I built it

I built Specters with Flask. Data is stored on a MongoDB Cluster, and during development, I used AI when I felt stuck trying to work it out by hand. Examples of times I used some AI are for quiz development, and some course rendering.

Challenges I ran into

A challenge I ran into was getting the site hosted since Vercel is serverless. It took me a while to find an old forum post telling me about file structures and file requirements. Also, in the showcase video, my mic was unable to be recorded.

Accomplishments that I'm proud of

I'm proud of the website looking and functioning good in the end. I'm also proud of it using Google Auth instead of a different auth solution.

What I learned

For this project, I learnt that Vercel is oddly specific with its requirements to host Flask sites. I also learnt how to use the Gemini API

What's next for Specters

My plan for Specters, is to turn it into my own ed-tech "startup" as it's been a dream of mine to create something innovative like this. I've already done some work for promoting Specters in my community, so my plan is to eventually make it profitable.