Megathread: Side Project Ideas - Scoped Out + Feasible

Image Converter

Context

A lot of the times, websites will have specific image requirements, often taking in only .png and .jpg. However, there are a lot of images out there that aren't in these formats, particularly the compressed .webp format when you download an image online (this is what Taro serves as well).

An organic use-case I've run into constantly is finding a good image for a lesson in a course I'm making, just to realize that it's .webp. Google Slides doesn't support .webp unfortunately, so I need to go convert it.

Build

This would be best as a web app. I can see this being a mobile app, but I have rarely run into this use-case when on my smart phone.

If you're a back-end engineer, you can also build this out by making an image conversion API:

• Input: Image resource (hosted at some CDN URL), targeted image file type
• Output: Converted image resource (again, hosted at some CDN URL)

Market Research

There are a bunch of websites like this one: https://cloudconvert.com/webp-to-png

They pretty much all have ads or account creation requirements (or both), which is annoying.

This is the classic product where you can simplify the interface, remove ads, and turn it into a simple utility app like I describe in my course (literally the Alex Playbook).

This use-case is small but common. These websites get millions of visits every single week. That cloudconvert site I've linked above has converted 2 billion+ files for users!

Find Middle Point Between 2 Locations

The main use-case I have run into here is to find a restaurant that is in between myself and a friend I am meeting. I'm a nice person and Bay Area traffic sucks, so I want the drive to be as fair and equal as possible for both parties.

Build

This could either be a mobile app or web-app. Since navigation lives mainly on smartphones now (Google Maps is such a good app!), it's probably better as a mobile app.

Market Research

Here's the biggest example I could find on Android: https://play.google.com/store/apps/details?id=com.cocozzello.tom.halfway&hl=en_US

The iOS one seems to be more popular (probably 50,000+ downloads?): https://apps.apple.com/us/app/whatshalfway/id1642536395

Here's some examples on web:

• https://www.meetways.com/
• https://www.whatshalfway.com/

Admittedly, this space doesn't seem to have the biggest demand. I can see something getting 10,000+ users in a year, but 100,000+ seems hard.

Make It Cool

This is a great product to make numerous optimizations to for the experience to be really minimal and beautiful:

• Pull your friend's address automatically from a contact book
• Address autocomplete (Google Locations API)
• Show detailed restaurant information (Google Places API) - Yelp API used to be great for this by being free for a long time, but it's quite expensive now 🥲

YouTube To mp3

A lot of folks don't have Spotify and use YouTube for music (I may or may not be one of them). Not only that, they rip the songs directly from the YouTube video.

Build

I have actually only seen this on web, so a web app would be great. I think it would be great as a mobile app, but I imagine Google is blocking every Android app trying to do this from Google Play. Apple is probably doing the same to keep things above board.

If you know how to develop native PC or Mac apps, that could be great too (I've seen this many times). However, it probably won't be hosted on the native stores. People are usually asking for folks to download the native apps from their website.

Market Reseach

There are a ton of websites out there that do this. Here are a couple:

• https://y2mate.nu/en-9le1/
• https://ytmp3.la/en-9CCY/

These websites get millions of visits every single week.

As always, these websites tend to be riddled with ads. They also have a lot of reliability problems, often bricking requests or hanging indefinitely. I imagine there's some meaty technical problems to solve here to guarantee product quality despite this being a Level 1 app.

Generate Subtitles From Video

When we get course videos from external creators, they often don't come with subtitles. This is an organic use-case that we run into constantly as Taro. Right now, we are doing hacky stuff where we upload the videos to Loom/YouTube and extract the generated subtitles from there. A focused tool here would be awesome.

A more organic use-case is you have some vacation videos, and you want your old, hard-of-hearing relatives to be able to understand what's going on in these videos.

Build

This is actually one of those projects where AI is pretty perfect. This could live as an LLM, a wrapper API, or a front-end application.

Market Research

There are a lot of websites that do this, but they're all paid/riddled with ads. It does make more sense with this tech as LLM costs are non-trivial:

• https://www.veed.io/tools/auto-subtitle-generator-online
• https://clipchamp.com/en/subtitle-generator/

I'm sure you could get a ton of users if you make this software free/ad-free, but it could end up costing you a lot of $$$, unlike my other ideas here.

One way you can avoid the cost issue is to make a module that moves the computation costs to the user. Here is a tool I've cloned off of GitHub which wraps OpenAI's Whisper: https://github.com/m1guelpf/auto-subtitle

The GitHub repo has 1,700+ GitHub stars, which I'm sure is enough to impress a lot of recruiters and hiring managers.

Pickleball ball comparison app

One idea is to build a simple utility app for something trendy that has a tailwind. The obvious one that is top-of-mind for me is, of course, pickleball. There are many people trying pickleball for the first time, and that number will increase in the coming years.

One thing that pickleball players do frequently is buy balls. Legit players will play with balls like Franklin, Dura, or Vulcan, but there are cheaper balls for beginners as well. Even within the top brands, there are differences

You could build a comparison table that shows the various attributes of the ball:

• Which competitions use it?
• How durable are they? Likelihood to break
• How cheap are they?

The obvious way to monetize this is to include affiliate links to purchase the balls. With a decent webapp, content marketing, and a few years of patience, this could generate meaningful revenue.