Posts
All the articles I've posted.
-
How to Use Perfectionism as a Tool — Discovering the Vertical Slice Development Approach
I tried to fix my perfectionism, but the attempt to fix it was itself perfectionism. So I changed direction. A record of redesigning my development process using 'vertical slices' — cutting vertically instead of stacking horizontally.
-
How Far Can AI Go in Researching User Pain? — Limits I Hit While Searching for a Portfolio Topic
In an age where AI lets you code fast, you get stuck at the stage of figuring out 'what to build.' Here are 4 ways to find people's pain points, and the limits I found when I handed that research over to AI.
-
PyTorch Image Classification Training Pipeline — An 11-Step Checklist
I abstracted the training flow I learned yesterday through the NDT hands-on exercise into a step-by-step checklist applicable to any image classification problem. Along with small questions like why normalization is needed.
-
NDT Defect Classification Hands-on — From PyTorch + ResNet18 Transfer Learning to Grad-CAM
An opportunity came up to put my embedded object recognition experience to use, so I picked up NDT (non-destructive testing) again. A hands-on record of training a 6-class defect classifier on the NEU Steel Defect dataset using transfer learning, and visualizing the model's decision basis with Grad-CAM.