classification

A one-word rewrite moving your accuracy several points is not a prompting problem, it is an uncalibrated decision function. The fix is measuring the model's prior over labels and dividing it out, not another round of tinkering.

Pushing every query through one pipeline wastes latency, money, and precision. A router decides what each query actually needs. The signal is naming the four decisions a router makes and how you stop a misroute from sinking the answer.

k-NN is quick to code and a solid check of vectorization and the lazy-learner tradeoff. Interviewers look for a tidy vectorized distance calculation plus knowing why naive prediction is O(n) and how to accelerate it. The implementation follows.

A from-scratch staple that checks whether you grasp the conditional-independence assumption and why you work in log space. The signal is fitting per-class means and variances, summing log-likelihoods, and adding the log prior. Here is the implementation.

SVMs sort the candidates who memorized 'maximize the margin' from those who can explain how a kernel delivers non-linear separation without ever touching the high-dimensional space. Here is the answer that lands the second signal.

Two simple classifiers that keep showing up in interviews because they check whether you grasp assumptions and tradeoffs, not just sklearn calls. What matters is k-NN's laziness and curse of dimensionality against Naive Bayes' independence assumption. Here is the answer.

LDA is the supervised cousin of PCA, and the contrast is a favorite. What matters is knowing that LDA uses the labels to maximize class separation while PCA only chases variance, plus the C minus 1 dimension cap that trips up most candidates.

Some models are multiclass out of the box; others need a wrapper to extend a binary classifier to K classes. What shows depth is knowing the three strategies and exactly when each one wins. Here is the answer.

Defaulting to 0.5 leaves money or safety on the table. The right cutoff comes from the cost of each error and the operating constraint, not the model. Here is how to set it deliberately.

Moderation is a multi-stage classification problem with harsh tradeoffs: false negatives cause real harm, false positives silence legitimate users, and the adversary keeps adapting. What interviewers reward: the tiered pipeline, per-severity precision/recall calibration, and human-in-the-loop. This lays out the design.

A high-AUC churn model that changes no behavior is worthless. The strong answer nails the label definition, chooses classification vs survival deliberately, and is judged on retention uplift, not accuracy.

Moderating billions of items a day with an LLM on each one is too slow and too expensive. Learn the tiered funnel (cheap classifier then LLM then human), how to tune thresholds for precision versus recall, and how to stay ahead of adversaries.

Toxicity detection looks like plain text classification but is nothing of the sort: context flips labels, adversaries evolve weekly, and naive models tag dialects as hate. The signal is naming those failure modes and building the human-in-the-loop system around them.