Case Studies

Review a multilingual classifier whose tokenizer fragments one language, an embedding search system whose nearest neighbors encode popularity rather than relevance, and a fine-tuned assistant that regresses on safety behavior.

Evidence and review

For each case issue a deploy, limit, or stop decision supported by an ablation. Claims must trace to versioned code and artifacts. Screenshots, unexplained generated prose, and a single favorable run do not pass.

Oral defense

Explain the mechanism, identify the most dangerous assumption, interpret one failure from raw evidence, and name the condition that would reverse your decision.

Source backbone

Use Dive into Deep Learning, PyTorch documentation, primary papers, and Designing Data-Intensive Applications selectively.