A WooCommerce recommendation engine scored 0.85 AUC in staging and fell apart on launch day because the live audience did not match the training data. Inverse probability weighting is how I got an honest read on what the model was really doing.
A client’s hourly sales model lost its accuracy every night at midnight because it read 23:00 and 00:00 as far apart. Sine and cosine encoding puts the hours back on a circle, which is where they belong.
A client bought a 64-core Threadripper and their script still ran on one core for an hour. Ray brought it down to about two minutes, with less code than the multiprocessing version I tried first.
Data work is moving away from generalists and toward people who know the domain, whether that is archaeology, renewable energy or sport. I got this wrong once on a solar reporting engine, and the fix was domain logic rather than a better query.
A client’s research pipeline kept hallucinating specs until we stopped writing one giant prompt. Notes on using IntelliNode’s VibeFlow to compile intent into a task graph, plus the code we run in production.
A WooCommerce site broke because its new host had no bcmath extension. The proposed filter for the required PHP extensions list in core would catch that kind of gap before checkout starts failing.
Predicting lead conversion from 800 WooCommerce rows, where XGBoost just memorized the noise. Why TabPFN and in-context learning made a better baseline than another afternoon of hyperparameter tuning.
A WooCommerce sync script went from ten minutes to three hours after the catalog grew. cProfile showed the API was fine and a nested loop building strings was eating 70% of the runtime.
A client’s detector found the gearboxes and ignored the bolts, and the problem was the loss function rather than the dataset. Here is how the YOLOv1 loss splits into five terms, why width and height get a square root, and what the lambdas are for.
MRR said a WooCommerce store search was fine while shoppers kept skipping the top result. Here is why binary relevance metrics mislead, and how NDCG and ERR score a result list closer to the way people read one.