My journey so far

I started taking side projects seriously early 2023 and built GoodWatch with the mindset of releasing something useful to the public that everyone can use for free.

My philosophy: if i build something that I need myself, others might find it useful too.

Fast forward to today and that mentality did not really change. I'm still releasing apps as open source that I like to use myself.

How's that a business?

It's not. I'm not earning any money with these. That's because most of my career choices are based on gut feeling. If it feels warm and like the right thing to do, I'm going with it. Fun & enjoyment is a huge part of my equation - never tried to learn things I'm not interested in, just because it's currently hyped in the job market.

So, my bet is building a personal brand. Eventually, I'll learn enough about distribution, marketing, social media and kicking my lack of will power to release something that gains a lot of traction.

You'll never make it without a monetization plan

Yeah, sure. I just can't and won't force it. I'll monetize when i feel like it. I'm getting closer. It looks like I first want the audience to pitch a real product.

How do you make money?

I'm earning 6-figures with my 9to5 as an engineering lead for an American company. Also earning almost 6-figures with my freelancing projects. It grew about 50% last year.

My customers:

• an international tax consultant for clients that trade with large volumes of investment funds (my biggest income)
• an american intelligent traffic tech company
• just recently added: an american robotics training company
• soon: small/mid-sized local companies that want me as an AI and automation consultant

Expanding my self-employment is what gives me the flexibility i thrive for.

My projects

Since financials are covered by the above, my side projects are a mixture of nerdy and really nerdy stuff. I'll briefly summarize here and write more about each individual work track in a series of posts.

👉 GoodWatch

Oldest project. Recommendations and data aggregator. Lots of features, but nothing good enough. Missing focus and lots of opportunities to grow.

Good: very solid foundation Meh: need to reprioritize work when i feel like it

👉 AI Stack

Started half a year ago and my current focus. It's about to become more community-focused and a learning hub. AI knowledge is so scattered across the web, that it needs an entry point, which is what I want to create.

Good: full-on Zeitgeist Meh: can't figure out the CLI/skill-sharing UX

👉 Coop

Current working title is "Together we build" and it's the first massive multiplayer coop game of its kind. Got tons of ideas, but very little experience with game design.

Good: Tech preview extremely promising Bad: Game mechanics are not fun so far

👉 Manaschmiede

A Magic the Gathering deck builder with an AI assistant. Building it because I want to print my dream decks and play them with my kids. Got the sleeves and a new printer with photo paper and now ready to do the first test run.

Good: proof of concept works just after 2 weekends Meh: can't validate quality of AI suggestions because i'm not a pro

👉 Immogrep

We (my wife and me) are looking for a new property as an investment. There are lots of portals and information is scattered around. Built an app that scrapes the web and consolidates all in one AirBNB-like map+list view. Helpful for keeping track with everything and getting notified whenever a new object is listed.

Good: The best filter options I ever saw on a real estate site Meh: hard to scrape sites with unstructured data

What's next?

I'll frequently share updates about anything on the above. Stay tuned for posts that go deeper into all kinds of challenges to ideate, build, motivate, market, etc.

Thanks guys, really value all of you ❤️

The Need For Pivot

I think I need to reflect on my journey with GoodWatch so far.

Why Pivot?

I have 300 registered users, 100 people in the Discord and about 100 daily visits. These are small numbers, but still a nice accomplishment.

The impact I'd like to have on the movie community is much bigger though. To achieve that, I need to pivot. I have ideas like:

• A tinder-like app to decide what to watch together when sitting on a couch
• A more social experience with people sharing their customized taste profiles
• A browser extension that enriches Netflix and other streaming sites with GoodWatch data (score from other sites, your taste match, etc.)

The way I'm building things for GoodWatch is kind of chaotic. I see an issue and I solve it. But I can't focus enough to do it the proper way. And some problems seem to be hard to crack (like SEO).

Where is this coming from?

I worked long on my last update. But it barely caused any reactions. Because it doesn't spark emotions. It's nothing that lets people feel joy and make them share it with their friends or followers. I understand that, I would feel the same way being a user.

This is not exactly demotivating, but lets me step back a bit and think about where I want to go with this project.

What's next?

I don't have any regrets. I put money into this project and it took a lot of time to get to this point. At the same time, I learned SO MUCH. I'm grateful for every piece of feedback I've gotten, and for every person reading my updates. It's genuinely my favorite way of working. Build in public is my thing.

For now, I'll keep GoodWatch in maintenance mode and work on stabilizing existing features (I know of a few bugs that need attention).

I will also focus on my other project <#1457348989321089220> which makes me excited because I think it solves a real problem for a lot of people. I might be wrong, but I'll validate it quickly and move on if traction is too low.

In the meantime, I'll explore which pivot for GoodWatch is the most promising and continue there.

Btw...

I'll start with video content for real. TikTok I'm coming. Soon.

Your Taste Profile

This is an important release. One of the biggest merge requests I ever pushed.

What's inside:

• Taste Detection based on many different attributes
• Personalized Recommendations
• Completely revamped Onboarding

Onboarding

Guests can try the Fingerprint & Recommendations even before signing up. The whole onboarding process has been redesigned from scratch. After 20 scores, it prompts the user to sign up to not lose their progress.

Taste

The more they score, the better and more personalized the profile gets:

• Genre map
• Time travel through decades
• Actors & Director preferences
• Favorite Fingerprint attributes

SEO

Fixed some problems with canonical tags where pages with query params were indexed as duplicates. That killed my Google results completely and will hopefully start to recover now.

Tech

This took so long because I had to figure out the stack that allows me to grab recommendations in real-time. The biggest win on that matter is Qdrant.

Also, I had to wait for my data pipeline to create a significant amount of fingerprint vectors for the engine to produce satisfying results.

Check it out here: https://goodwatch.app/taste/quiz