Training Data
322 episodes · Page 6 of 14
#5402: Amazon Go and the Humans Behind the AI
Amazon promised a store with no checkout. The Information reported 700 of every 1,000 sales needed human review in 2022.
#5400: Amy, Presto, and the Thousand Workers Behind "AI
From X.AI's Amy to Amazon's Just Walk Out, the humans quietly doing the "intelligent" part while the product says AI.
#5396: Teaching a Small Model to Stop Spelling Out Numbers
Your ASR pipeline is fine until someone dictates "three point two" and gets "three point two" spelled out. Here's how inverse text normalization ac...
#5393: Fine-Tuning a Model on 100 Hand-Edited Answers
You don't need 10,000 examples to make a model sound like you. The real number is closer to 100 — if the edits are opinionated.
#5389: Hugging Face vs Kaggle: Where Models Actually Live
Hugging Face and Kaggle aren't rivals — one is infrastructure, one is a practice field. Here's how the two platforms actually differ.
#5270: Teardowns as Documentation: Inside the Hardware Repair Playbook
A dead Dremel sparks a deep dive into teardowns as a documentation discipline — and why board-level knowledge matters more than ever.
#5256: Ruby Won the AI Benchmark. Is It Dying Anyway?
97% of public Ruby repos sat untouched for six months — yet Ruby beat Python in an AI coding benchmark. What's actually declining?
#5245: Training a Model on 100 Phone Photos
One drawer of screws, a hundred phone photos, and a schema. When does a small custom model beat a fine-tuned VLM?
#5199: Shell Companies Over Shell Companies
Why covert networks stack shell companies in layers — and why the structure itself is the weapon.
#5198: Counting the AI Models Nobody Releases
YOLO is free — the labeled data isn't. Why the custom image models that matter most are the ones you'll never see on Hugging Face.
#5187: Translation Is Not a Pipe: What Machines Miss
Rubrics rate machine translation above human work — until readers pick side-by-side. What translators know that metrics can't see.
#5169: Fine-Tuning vs From-Scratch for Minor Languages
What 774 experiments and a new Armenian model reveal about the tradeoff between fluency and knowledge in low-resource languages.
#5118: Model Identity Fraud: Distillation or Data Contamination?
When AI models lie about who they are, is it stolen knowledge or just contaminated training data?
#5117: How AI Training Data Gets Filtered (and Exploited)
Six stages of content filtering stand between raw web crawls and your AI model — here's where poisoning attacks slip through.
#5116: Beyond Common Crawl: The Datasets Shaping AI
Common Crawl is just the bulk flour. We explore the curated datasets that actually shape what AI models know and do.
#5113: Can Rival Labs Poison AI Training Data?
A few hundred crafted documents can shift a model's stance. Here's how data poisoning actually works.
#5019: Designing a Kosher Korean BBQ Restaurant
What happens when K-drama food culture meets kashrut? We design a restaurant that bridges both worlds.
#4971: Fine-Tune vs Prompt: Taming Hallucinated Deadlines
Big models add fake deadlines. Small models miss tasks. Which wrongness can you build a system around?
#4970: Labels vs. Book Covers: Structured Extraction Tradeoffs
General vision models vs. purpose-built pipelines for extracting data from messy real-world labels and covers.
#4969: How to Train a One-Class Detector from 300 Photos
From photo folder to production YOLO model — the real walkthrough for custom object detection.
#4809: Can a Podcast Train Itself?
Can a podcast become a self-developing project? We explore RAG vs. fine-tuning for lore memory.
#4788: How Weather Prediction Became a Science
From hand calculations to chaos theory to machine learning — the full arc of forecasting.
#4775: The Evaluator Role No One's Building For
Benchmarks like MMLU are broken. A new role is emerging: the domain-specific AI evaluator.
#4774: When a Token Isn't a Word Anymore
Images, audio, and code all use "tokens" — but they're completely different things. Here's how tokenization actually works across modalities.