Skip to content

amitgambhir/ai-terms-and-myths

Repository files navigation

📖 AI Terms & Myths

A practitioner's reference for AI/ML terminology and the misconceptions that won't die.

Markdown License Contributions Welcome

Browse the site — searchable, with dark mode


The Problem

AI/ML conversations are drowning in vague terminology and confident misconceptions. Engineers hear "just fine-tune it" without understanding what that means mechanically. Decision-makers assume more parameters equals smarter. Teams ship RAG systems without knowing what hallucination actually is.

The result: bad architectural decisions, wasted compute, and products built on assumptions that sound right but aren't.


What It Does

Two reference documents, maintained by practitioners, designed to be read when you need a precise answer — not a tutorial, not a textbook, not a blog post.

Document What's Inside Scale
ai-terms-glossary.md 80+ AI/ML terms defined with precision and practical context A–Z, alphabetized
ai-myths-busted.md 20+ common misconceptions debunked with evidence and consequences Each tied to real decisions

Every entry follows a strict format. Glossary terms are self-contained 3–6 sentence definitions. Myths include a Reality section with named papers, models, or techniques — and a Why it matters section that connects to decisions you actually make.


Demo

Glossary entry:

### Retrieval-Augmented Generation (RAG)
A technique where an LLM is given relevant documents retrieved from an external
knowledge base before generating a response. The model does not "know" the
retrieved information — it uses it as in-context evidence for that specific query.
RAG reduces hallucination by grounding generation in source material, but does
not eliminate it.

Myth entry:

## "More parameters means a smarter model"

Reality: A smaller model trained on high-quality data with good techniques can
significantly outperform a much larger model trained on poor data. The Chinchilla
paper demonstrated that most large models were over-parameterized relative to
their training data. Microsoft's Phi-2 (2.7B parameters) outperformed models
many times its size on several benchmarks simply due to better data curation.

Why it matters: Default to the model that fits your task, latency, and cost
requirements — not the largest one available.

Built On

  • Markdown — plain text, no build step, readable anywhere, diffs cleanly in PRs
  • GitHub Issues + Templates — structured contribution workflow with pre-flight checklists
  • Community review — every entry is validated for accuracy, tone, and practical relevance before merge
  • MkDocs Material — documentation site with built-in search, dark/light toggle, auto-deployed via GitHub Actions

Quickstart

git clone https://github.com/amitgambhir/ai-terms-and-myths
cd ai-terms-and-myths

Open ai-terms-glossary.md or ai-myths-busted.md in any markdown viewer. That's it.


Contributing

Every contribution goes through an issue-first workflow:

  1. Open an issue using the templates in .github/ISSUE_TEMPLATE/
    • Glossary terms: prefix with [GLOSSARY] Suggest: [Term Name]
    • Myths: prefix with [MYTH] Debunk: "[Misconception]"
  2. Wait for feedback before writing a PR
  3. Match the tone and density of neighboring entries
  4. Run through the precision checklist in CONTRIBUTE.md

What gets in

  • Terms that appear in research papers, framework docs, or serious industry discussion
  • Myths that are widely believed and change real decisions when wrong
  • Entries where every claim holds up under scrutiny

What doesn't

  • Marketing jargon, straw men, or trivial facts
  • Entries that duplicate existing content
  • Anything that ages poorly (transient news, market claims)

Why This Is Different

  • No filler — banned words include "leverage", "unlock", "seamlessly", "robust", and "empower"
  • Practitioner-first — every entry is written for people building systems, not writing papers
  • Evidence-backed — myths cite specific papers, models, and techniques by name — no "some research shows"

Built for the people who need to know what the words actually mean before they ship.

About

A practitioner's reference for AI/ML terminology and the misconceptions that won't die. 80+ precise glossary definitions and 20+ myth-busting entries — no jargon inflation, no hand-waving.

Topics

Resources

Contributing

Stars

Watchers

Forks

Packages

 
 
 

Contributors