Build journal — actively updating

Figuring it out.One journey at a time.

A developer's honest record of building things — products, software, hardware — without a roadmap. Documented in public so you can follow along, steal ideas, or just watch what happens.

Keychains
From a five-year-old idea to a product worth selling. · 1 posts
Building
🏒
Goal Light
Teaching a computer to watch hockey. · 3 posts
Active
Next journey — TBD
Something is always in the pipeline
Planned
Active journeys All journeys
Product build 01
Keychains
From a five-year-old idea to a product worth selling.

Custom car keychains in stainless steel with CNC walnut wall holders. The idea sat in my head for years. This is the attempt to actually build it.

Prototype phase 5%
ML project 02
🏒
Goal Light
Teaching a computer to watch hockey.

A machine learning system that detects when a goal is scored on a live broadcast and triggers lights behind my TV. Started with the Leafs, being built to work for any team.

Training data phase 35%
Latest posts All posts
Goal Light Build log
How to extract every Nth frame from a video for ML training

60 videos. 210,000 frames. About 0.5% of them matter. This is how I got the frames out — and why I'm only saving a 400×100 pixel crop of each one.

Goal Light Build log
How to download YouTube videos for ML training data

Before you can train a model you need data. Before you have data you need footage. This is how I got it and why highlight reels beat full broadcasts for this problem.

Keychains Build log
I had this idea for years. It started with a car I had no business loving.

The keychain concept started with a car and a feeling that everything in the automotive accessories market was either cheap or soulless.

Tutorial Tutorial
How to set up a Python virtual environment properly — and why it matters

The one thing you should do before every ML project. Takes 30 seconds and saves hours of dependency hell later.

Goal Light Build log
Deciding what to build and why

I wanted my living room to react when the Leafs score. Here's the full system design and why I chose computer vision over an API.

Goal light — step by step Full journey
01 Deciding what to build and why
Build log
02 How to download YouTube videos for ML training data
Build log
03 How to extract every Nth frame from a video for ML training
Build log
04 Training the model — coming soon
In progress

No spam. Unsubscribe any time.