Resource Library
All learning materials organized by type and status.
96 resources
Filters
All Resources
What is AI
Notes on AI
video
Notes on AI
Most arguments about AI are actually arguments about **definitions**. The term "AI" is broad and vague, covering everything from simple spell-checkers to sci-fi killer robots.
Links to: 2 nodes
Why Generative AI Feels Different
Notes on AI
video
Notes on AI
Traditional AI (Machine Learning) has existed for decades—classifying data, ranking search results, and predicting outcomes. While useful, it always felt like a "system" or an algorithm. It was mec...
Links to: 3 nodes
How AI Thinks
Notes on AI
video
Notes on AI
In this note, we explore the mechanism of "thought" in Generative AI. We focus on Large Language Models (LLMs) as the core example.
Links to: 3 nodes
Why GenAI Advanced All at Once
Notes on AI
video
Notes on AI
We explore why Generative AI seemed to advance across all fronts—text, image, audio, video—simultaneously.
Links to: 3 nodes
What Is a Model, Really?
Notes on AI
video
Notes on AI
Reframe a model as compressed patterns, not understanding or knowledge.
Links to: 4 nodes
Training vs Using a Model
Notes on AI
video
Notes on AI
*Derived from the Transcript.*
Links to: 3 nodes
What AI Receives When You Send a Prompt
Notes on AI
video
Notes on AI
A **Prompt** is commonly misunderstood as the sole input to an AI model. In reality, it is only the visible "Top Slice" of a larger input stack, best understood as a **Prompt Sandwich**.
Links to: 3 nodes
Is AI a Student or an Actor
Notes on AI
video
Notes on AI
We often mistake AI outputs for fact-checked answers from a reasoning process. In reality, AI generates **completions**—continuations of patterns, similar to an **Improvisational Actor** following ...
Links to: 4 nodes
What AI Is Good At vs Bad At
Notes on AI
video
Notes on AI
This episode introduces a practical framework for understanding when to use AI by framing it as a "special employee" with three distinct characteristics.
Links to: 4 nodes
The 5-Sentence Mental Model of GenAI
Notes on AI
video
Notes on AI
This episode provides a checkpoint after the foundational episodes, compressing the key concepts into five memorable sentences that serve as a mental compass for AI.
Links to: 1 node
Tokens
Notes on AI
video
Notes on AI
AI models do not read words. They read **tokens** — the basic unit of text a model processes. A token is close to a word but not the same: one word can be one token, several tokens, or several word...
Links to: 5 nodes
Tokenization
Notes on AI
video
Notes on AI
**Tokenization** is the process of turning raw text into tokens before an AI model processes it. It is **preprocessing**, not thinking: the model only sees the resulting pieces. Tokenization is **l...
Links to: 4 nodes
Why Typos Don't Matter
Notes on AI
video
Notes on AI
**Why Typos Don't Matter** explains that AI models don't read words - they read tokens, small chunks of characters. When you misspell a word like "understanding" as "undersatnding", most of the tok...
Links to: 2 nodes
Context Window
Notes on AI
video
Notes on AI
**Context Window** is the amount of information a model can see at one time. It's not memory - it's working space. The model can only reason about what is currently visible in its context window. A...
Links to: 4 nodes
Context Engineering
Notes on AI
video
Notes on AI
**Context Engineering** is the practice of shaping the information environment the model operates in, not just writing better prompts. The prompt is not what the model responds to - it responds to ...
Links to: 3 nodes
Why Long Chats Drift
Notes on AI
video
Coming soon
Notes on AI
Explain how bloated context and topic mixing degrade outputs.
“Forgetting” vs “Never Knew”
Notes on AI
video
Coming soon
Notes on AI
Distinguish two different failure modes that users often confuse.
System vs User Messages
Notes on AI
video
Coming soon
Notes on AI
Explain how different message roles steer model behavior.
Why AI Sounds Confident
Notes on AI
video
Coming soon
Notes on AI
Expose fluency bias as the reason confident answers can still be wrong.
**Why Chatbots Forget — And Why RAG Exists**
Notes on AI
video
Coming soon
Notes on AI
Introduce RAG as a response to context and memory limits.
Grounding: Answer Only From My Text
Notes on AI
video
Coming soon
Notes on AI
Show how grounding reduces hallucinations by constraining sources.
Synthesis: How to Keep AI Sharp
Notes on AI
video
Coming soon
Notes on AI
Summarize practical habits for maintaining output quality.
Asking for Assumptions
Notes on AI
video
Coming soon
Notes on AI
Teach how surfacing assumptions reduces wrong answers.
Output Formats
Notes on AI
video
Coming soon
Notes on AI
Show how structure improves reliability and usefulness.
Quick Verification Habits
Notes on AI
video
Coming soon
Notes on AI
Provide lightweight ways to sanity-check outputs.
Mini-Synthesis: The Trust Ladder
Notes on AI
video
Coming soon
Notes on AI
Introduce a staged model for trusting AI outputs.
The 3-Part Prompt
Notes on AI
video
Coming soon
Notes on AI
Teach a repeatable prompt structure that works across tasks.
Few-Shot Prompting
Notes on AI
video
Coming soon
Notes on AI
Explain how examples teach behavior without retraining.
Stepwise Prompting
Notes on AI
video
Coming soon
Notes on AI
Show how breaking work into steps improves quality.
“Critique Then Rewrite”
Notes on AI
video
Coming soon
Notes on AI
Demonstrate iterative improvement as a core AI skill.
“Ask Me Questions First”
Notes on AI
video
Coming soon
Notes on AI
Explain when and why models should clarify before answering.
Tone Control Without Cringe
Notes on AI
video
Coming soon
Notes on AI
Show how to steer tone explicitly and professionally.
What Hallucinations Are
Notes on AI
video
Coming soon
Notes on AI
Define hallucinations accurately without mystique.
When You Should Trust AI
Notes on AI
video
Coming soon
Notes on AI
Build intuition for low-risk vs high-risk use cases.
Red Flags
Notes on AI
video
Coming soon
Notes on AI
Teach fast signals that outputs may be unreliable.
Calibration
Notes on AI
video
Coming soon
Notes on AI
Show how to ask about confidence without false precision.
Asking for Sources
Notes on AI
video
Coming soon
Notes on AI
Explain what sources mean in AI outputs and their limits.
Contradictions
Notes on AI
video
Coming soon
Notes on AI
Teach how to spot internal inconsistencies quickly.
**Prompt Injection**
Notes on AI
video
Coming soon
Notes on AI
Introduce prompt injection as a real-world security risk.
Verification in 30 Seconds
Notes on AI
video
Coming soon
Notes on AI
Provide a minimal verification workflow.
Reliability vs Creativity
Notes on AI
video
Coming soon
Notes on AI
Explain tradeoffs between consistency and originality.
Synthesis: From Brainstorming to Decisions
Notes on AI
video
Coming soon
Notes on AI
Tie trust concepts into a usable decision workflow.
Training vs Usage Revisited
Notes on AI
video
Coming soon
Notes on AI
Reinforce why models do not learn from conversations.
Pretraining
Notes on AI
video
Coming soon
Notes on AI
Explain how models learn general patterns at scale.
Instruction Tuning
Notes on AI
video
Coming soon
Notes on AI
Show why chat models follow instructions.
Preference Training
Notes on AI
video
Coming soon
Notes on AI
Explain why models feel helpful and polite.
Fine-Tuning
Notes on AI
video
Coming soon
Notes on AI
Clarify when fine-tuning helps and when it misleads.
Knowledge Cutoff
Notes on AI
video
Coming soon
Notes on AI
Explain why models lack recent information.
Memorization vs Generalization
Notes on AI
video
Coming soon
Notes on AI
Distinguish recall from pattern learning.
Overfitting
Notes on AI
video
Coming soon
Notes on AI
Explain overfitting with simple intuition.
Benchmarks
Notes on AI
video
Coming soon
Notes on AI
Show what benchmarks reveal and what they hide.
Synthesis: The Training Pipeline
Notes on AI
video
Coming soon
Notes on AI
Present the full training lifecycle in one mental picture.
Why Models Don’t Know Your Company
Notes on AI
video
Coming soon
Notes on AI
Explain why private knowledge is inaccessible by default.
**Agents Explained Conceptually**
Notes on AI
video
Coming soon
Notes on AI
Introduce agents as loops of models, tools, and memory.
Tools vs Models
Notes on AI
video
Coming soon
Notes on AI
Clarify division of labor between reasoning and action.
Memory Illusions
Notes on AI
video
Coming soon
Notes on AI
Explain why chat history feels like memory but is not.
Mini-Synthesis: Knowledge vs Behavior
Notes on AI
video
Coming soon
Notes on AI
Separate what models know from how they behave.
Input vs Output Tokens
Notes on AI
video
Coming soon
Notes on AI
Explain how token flow drives cost.
Why System Prompts Cost Too
Notes on AI
video
Coming soon
Notes on AI
Reveal hidden cost sources in AI interactions.
Context Is Expensive
Notes on AI
video
Coming soon
Notes on AI
Explain why longer context increases cost and latency.
Cost Hygiene
Notes on AI
video
Coming soon
Notes on AI
Teach habits that reduce AI costs.
Small vs Big Models
Notes on AI
video
Coming soon
Notes on AI
Explain routing decisions between models.
Speed, Quality, Cost Triangle
Notes on AI
video
Coming soon
Notes on AI
Present the fundamental tradeoff in AI systems.
Rate Limits
Notes on AI
video
Coming soon
Notes on AI
Explain why systems throttle or fail unexpectedly.
Caching
Notes on AI
video
Coming soon
Notes on AI
Show how reuse reduces cost and latency.
Free Trial Economics
Notes on AI
video
Coming soon
Notes on AI
Explain why AI business models changed.
Sensitive Data
Notes on AI
video
Coming soon
Notes on AI
Define what should never be shared with AI.
Personal vs Business Data
Notes on AI
video
Coming soon
Notes on AI
Distinguish risk levels of different data types.
Do Not Paste Secrets
Notes on AI
video
Coming soon
Notes on AI
Explain safer alternatives to sharing secrets.
Data Retention
Notes on AI
video
Coming soon
Notes on AI
Explain what data is typically stored and why.
Consumer vs Enterprise AI
Notes on AI
video
Coming soon
Notes on AI
Explain why organizations impose stricter rules.
Prompt Injection Revisited
Notes on AI
video
Coming soon
Notes on AI
Show real-world examples of prompt attacks.
Redaction
Notes on AI
video
Coming soon
Notes on AI
Teach how to remove identifiers without losing meaning.
Synthesis: AI Safety Policy
Notes on AI
video
Coming soon
Notes on AI
Summarize practical personal safety rules.
Vision Models
Notes on AI
video
Coming soon
Notes on AI
Explain what seeing means for AI systems.
OCR vs Vision Reasoning
Notes on AI
video
Coming soon
Notes on AI
Clarify why screenshots confuse models.
Image Generation Basics
Notes on AI
video
Coming soon
Notes on AI
Explain how images are generated from noise.
Inpainting
Notes on AI
video
Coming soon
Notes on AI
Show how partial image editing works.
Deepfakes 101
Notes on AI
video
Coming soon
Notes on AI
Explain what is easy and what is hard today.
Speech-to-Text
Notes on AI
video
Coming soon
Notes on AI
Explain factors affecting transcription accuracy.
Text-to-Speech
Notes on AI
video
Coming soon
Notes on AI
Explain tradeoffs between naturalness and control.
Multimodal Prompting
Notes on AI
video
Coming soon
Notes on AI
Show how to combine text, image, and audio inputs.
Why Chatbots Forget Revisited
Notes on AI
video
Coming soon
Notes on AI
Reconnect memory limits to system design.
Embeddings Explained
Notes on AI
video
Coming soon
Notes on AI
Explain semantic meaning as coordinates.
Chunking
Notes on AI
video
Coming soon
Notes on AI
Explain why document splitting matters.
Vector Databases
Notes on AI
video
Coming soon
Notes on AI
Explain what vector stores actually store.
Retrieval vs Generation
Notes on AI
video
Coming soon
Notes on AI
Separate evidence retrieval from text generation.
Reranking
Notes on AI
video
Coming soon
Notes on AI
Explain how the best evidence is selected.
Citations
Notes on AI
video
Coming soon
Notes on AI
Explain how grounded answers are produced.
Freshness Without Retraining
Notes on AI
video
Coming soon
Notes on AI
Explain how systems update knowledge safely.
Evaluating RAG
Notes on AI
video
Coming soon
Notes on AI
Explain how to test retrieval quality.
Synthesis: Build a Knowledge Assistant
Notes on AI
video
Coming soon
Notes on AI
Tie all components into a single system blueprint.
Acceleration in AI Solving Erdős Problems
Thoughts on AI
video
Thoughts on AI
Observing the unprecedented acceleration of AI systems solving previously unsolvable Erdős problems in January 2026, signaling a potential breakthrough moment.
Why Blockchain and LLMs Are More Similar Than You Think
Thoughts on AI
video
Thoughts on AI
Exploring the surprising structural similarity between blockchains and LLMs — both are closed systems that cannot see the real world by themselves.
The Oracle Problem and the AI Grounding Problem
Thoughts on AI
video
Coming soon
Thoughts on AI
Drawing the parallel between blockchain's Oracle Problem and AI's grounding challenge — both face the same fundamental trust issue when connecting to reality.
Trust — The Verification Chain
Thoughts on AI
video
Coming soon
Thoughts on AI
Most AI failures are not intelligence failures — they are trust and verification failures. The future is not just smarter models, but better bridges and better checks.