The Best Way to Read a Book (That Nobody's Doing)

⬅️ Back to Videos

▶️ Watch the Video

📝 VIDEO INFORMATION

• Content Type: Demonstration / Tutorial / Discussion
• Title: “The Best Way to Read a Book (That Nobody’s Doing)”
• Creator(s): Jeremy Howard (Demonstrator), Eric Ries (Author/Discussant), Jon O. (Moderator)
• Platform: YouTube
• Duration: ~43 minutes
• Publication Date: January 20, 2026
• Link: https://www.youtube.com/watch?v=zIqLuuyxgE4

E-E-A-T Assessment

• Experience: 5/5 - Jeremy Howard is the founder of fast.ai, a leading AI education platform, and has been at the forefront of practical AI application. Eric Ries is the author of “The Lean Startup” and the book being read. Both have deep, hands-on experience with their respective domains.
• Expertise: 5/5 - Jeremy demonstrates sophisticated understanding of LLM capabilities, context management, and workflow design. His approach shows mastery of AI-assisted learning at the highest level.
• Authoritativeness: 5/5 - Jeremy is a recognized leader in AI education and practical AI application. Eric Ries is a bestselling author and thought leader in entrepreneurship. Their combined perspective carries significant weight.
• Trust: 5/5 - Both speakers are transparent about their process, limitations, and the experimental nature of the approach. Jeremy openly shares his skepticism and verification process. No commercial agenda beyond genuine knowledge sharing.

Verdict: Proceed with review - This is a primary source demonstration from practitioners operating at the cutting edge of AI-assisted learning. The transparency, specificity, and grounded nature of the content makes it highly credible.

🎯 HOOK

What if you could read a book with a research assistant who knows everything, never gets tired, and can explore any rabbit hole instantly? Jeremy Howard demonstrates a revolutionary approach to close-reading that transforms a book from a linear consumption experience into an interactive dialogue. Using an LLM as a collaborative partner, he shows how to engineer context, verify claims skeptically, follow curiosity wherever it leads, and apply insights to your own situation-a method that was literally impossible just months ago.

💡 ONE-SENTENCE TAKEAWAY

The future of deep reading belongs to those who treat LLMs not as summarization tools but as collaborative partners requiring careful context engineering, enabling skeptical verification, curiosity-driven exploration, and personalized application that transforms how we engage with complex ideas.

⚖️ VERDICT

Overall Rating: 10/10

This video is a masterclass in using LLMs for deep learning and represents a genuine paradigm shift in how we can engage with written material. Jeremy doesn’t just explain-he demonstrates a complete workflow in real-time, showing exactly how to set up context, maintain continuity across sessions, verify claims skeptically, and explore rabbit holes. The conversation with Eric Ries provides meta-commentary on the reading experience that adds depth and validation. This isn’t speculative or theoretical; it’s a concrete demonstration of a methodology that viewers can immediately adopt. The only limitation is that it assumes some technical comfort with LLMs, though the principles are accessible to anyone willing to invest the setup time.

📊 EVALUATION CRITERIA

📖 SUMMARY

This video captures Jeremy Howard demonstrating a revolutionary approach to reading complex books using large language models. Working through Chapter 1 of Eric Ries’s latest book, Jeremy shows how to transform reading from passive consumption into an active, exploratory dialogue with an AI assistant that has been carefully prepared with comprehensive context.

The core insight is that LLMs enable a form of “close-reading” previously reserved for humanities scholars with years of erudition-except now anyone can access this capability with proper context engineering. Jeremy’s approach involves creating hierarchical summaries (chapter summaries, part summaries, whole book summaries), adding personal context (his discussions with Eric, his company’s launch post), and using the LLM as a skeptical, curious reading partner who can explain references, verify claims, and explore connections.

What makes this demonstration exceptional is its specificity. Jeremy shows his actual prompts, his actual LLM conversations, and his actual workflow. He doesn’t just claim this works-he proves it by reading through Chapter 1 in real-time, asking skeptical questions, following rabbit holes (the connection between FedMart’s Sol Price and Amazon, the Boeing/GE management lineage, the Bentham/Thompson philosophical contrast), and verifying specific claims by checking footnotes.

The conversation with Eric Ries provides crucial meta-commentary. Eric acknowledges that while he knew context matters with LLMs, he was surprised by the depth of Jeremy’s preparatory work. This two-hour investment in setup transforms the reading experience from scanning for main ideas (Jeremy’s usual approach to business books) to hours of deep engagement with fascinating tangents and historical nuance.

Jeremy’s method also addresses a fundamental challenge with complex books: they contain more rabbit holes than any single reader can pursue. By using an LLM as a research partner, he can follow his curiosity wherever it leads-checking historical claims, exploring philosophical connections, questioning whether something is “greed” or “different incentives,” and getting context on references he doesn’t recognize.

The video concludes with Jeremy’s process for maintaining continuity across reading sessions (handoff notes for “future me”), demonstrating how to treat each new chapter as a fresh LLM session while preserving accumulated context. This practical detail makes the methodology immediately actionable for viewers.

What the Video Covers

The demonstration follows a clear structure:

(00:00) Context Engineering - Jeremy shows his setup process: creating hierarchical summaries of the book (chapter, part, whole book), adding personal context (discussions with Eric, launch post), and preparing footnotes for easy reference.

(02:30) Hierarchical Context - Demonstrating how to create summaries at multiple levels of abstraction, allowing the LLM to understand both the details of the current chapter and how it fits into the larger work.

(07:00) Curiosity-Driven Exploration - Reading Chapter 1 on FedMart’s Sol Price, Jeremy explores rabbit holes: the Sol Price/Amazon connection, the “silent monitor” system, the Bentham/Thompson philosophical contrast, and contemporary accounts of historical events.

(12:00) Personalized Application - Jeremy applies the book’s framework to his own situation as CEO of Answer AI, using the LLM to explore how concepts like “ethos” and “financial gravity” might apply to his company.

(17:30) Following the Thread - Deep exploration of the Boeing/GE management story, following connections between Welch disciples, corporate culture destruction, and the 737 MAX disaster. The LLM helps trace these connections and provides historical context.

(25:00) Skeptical Verification - Jeremy demonstrates his skeptical reading approach: checking specific figures, questioning whether something is “editorializing,” and asking the LLM to verify claims and identify what exactly is being argued.

(33:30) Good-Faith Skepticism - Eric and Jeremy discuss how Jeremy combines skepticism with good faith-being willing to push back but also willing to be convinced when presented with additional information.

(35:00) Session Continuity - Jeremy shows his process for maintaining context across reading sessions using “handoff notes” that explain everything a “new” LLM instance needs to know to continue the reading partnership.

(39:30) Concept Exploration - Deep dive into “financial gravity” and evolutionary psychology concepts, using the LLM to explain academic literature and connect it to practical business situations.

(42:30) Meta-Reflection - Eric and Jeremy reflect on the experience, discussing how this approach enables a level of close-reading previously requiring years of erudition, and how LLMs might democratize access to deep engagement with complex texts.

Who Created It & Why It Matters

This video features three key contributors:

Jeremy Howard is the primary demonstrator-the founder of fast.ai, a leading AI education platform, and a recognized expert in practical AI application. His credentials are exceptional: he’s taught hundreds of thousands of people to use AI effectively, created popular open-source libraries, and consistently operates at the frontier of what’s possible with LLMs. What makes Jeremy’s perspective unique is his combination of technical expertise with pedagogical clarity-he doesn’t just use these tools, he teaches others how to use them.

Eric Ries is the author of the book being read and a discussant. As the creator of the Lean Startup methodology and author of one of the most influential business books of the past decade, Eric provides crucial meta-commentary on how his work is being received and interpreted. His reactions-surprise at the depth of context preparation, appreciation for Jeremy’s skeptical approach, reflections on close-reading as a technology-add layers of meaning to the demonstration.

Jon O. (presumably Jon O’Keefe or similar) moderates the discussion, keeping the conversation moving and ensuring key points are addressed.

The video matters because it moves beyond abstract speculation about AI’s impact on learning to a concrete, replicable methodology. Jeremy’s demonstration is specific enough that viewers can attempt to replicate it immediately. The conversation addresses both the technical how-to and the deeper philosophical implications of what it means to read deeply in an age of AI assistance.

Core Argument & Evidence

The central thesis is that LLMs enable a new form of close-reading that combines the depth of humanities scholarship with the accessibility of modern technology-but only with proper context engineering and workflow design.

Evidence supporting this argument:

1. Concrete demonstration: Jeremy shows his actual workflow, prompts, and LLM conversations in real-time
2. Specific examples: Multiple instances of rabbit holes explored (Boeing/GE, Bentham/Thompson, Sol Price/Amazon)
3. Meta-validation: Eric Ries confirms that this approach extracts more value from his book than traditional reading
4. Comparative baseline: Jeremy explicitly states his usual approach to business books (scanning for main ideas) and contrasts it with this deeper engagement
5. Historical parallel: The comparison to humanities close-reading requiring years of erudition establishes what this method democratizes

Logical structure:

• Premise 1: Traditional close-reading requires extensive background knowledge and time investment
• Premise 2: LLMs can provide context, verify claims, and explore connections on demand
• Premise 3: But LLMs require careful context engineering to be effective partners
• Premise 4: When properly set up, LLMs enable skeptical, curious, deep engagement with complex texts
• Conclusion: This transforms reading from consumption into collaborative exploration

The argument is compelling because it’s demonstrated rather than asserted. Jeremy doesn’t claim this works-he shows it working, in real-time, with a specific book and specific questions.

Practical Applications

For readers of complex books:

• Invest 2 hours in context engineering (hierarchical summaries, personal context, footnotes) before starting
• Use the LLM as a skeptical partner-ask “is this true?” and “what am I missing?”
• Follow rabbit holes that interest you, treating the LLM as a research assistant
• Apply concepts to your own situation by sharing your context with the LLM
• Maintain continuity with handoff notes between reading sessions

For authors of complex books:

• Recognize that readers may engage with your work at unprecedented depth using LLMs
• Consider how your references and allusions will be received by AI-assisted readers
• Appreciate that skeptical verification becomes easier, raising the bar for accuracy
• Understand that readers can now follow connections you only hint at

For educators:

• This methodology can be taught to students as a new form of literacy
• The approach combines traditional close-reading with modern AI assistance
• Context engineering becomes a crucial skill for deep learning
• Students can explore primary sources with AI assistance that was previously unavailable

🔍 INSIGHTS

Core Insights

• Context Engineering is the Critical Skill: Jeremy’s two-hour investment in setting up context (hierarchical summaries, personal background, footnotes) is what makes the subsequent reading experience transformative. The tools don’t matter as much as the preparation. As Eric notes, “It’s a lot more effort than I would have put into preparatory work”-but that preparation is what enables everything else.

• The Shift from Consumption to Dialogue: Traditional reading is linear consumption. Jeremy’s approach treats the book as the starting point for an exploratory dialogue. The LLM becomes a research partner who can explain references, verify claims, explore historical connections, and challenge interpretations. This transforms reading from passive to active.

• Skeptical Verification Becomes Standard: Jeremy repeatedly fact-checks, questions whether something is “editorializing,” and asks for evidence. This isn’t adversarial-it’s rigorous engagement. The LLM enables readers to maintain healthy skepticism without getting stuck, because verification is instant and contextual.

• Democratization of Close-Reading: Eric notes that traditional close-reading required “years and years and years of study” to understand references and context. LLMs democratize this capability, allowing anyone to engage deeply with complex texts that assume extensive background knowledge.

• Rabbit Holes as Feature, Not Bug: Jeremy explicitly states this approach takes “dramatically longer” than his usual book-reading because there are “so many rabbit holes I’m finding that are fascinating.” In traditional reading, this would be inefficient. With LLM assistance, following curiosity wherever it leads becomes viable.

• The Attention Economy of Context: LLMs have limited context windows, making context management critical. Jeremy’s approach of hierarchical summaries (chapter → part → whole book) is a solution to this constraint, providing multiple levels of abstraction within context limits.

• Continuity Through Handoff Notes: The practical solution to session boundaries-having the LLM write detailed notes for “future me”-turns a technical limitation into an opportunity for reflection and synthesis.

• Good-Faith Skepticism: The ideal approach combines skepticism (willingness to question) with good faith (willingness to be convinced). Jeremy demonstrates this repeatedly-pushing back on claims, but accepting explanations when they’re supported.

How This Connects to Broader Trends/Topics

• The Decade of Agents: This exemplifies Andrej Karpathy’s framing of our era as “the decade of agents.” Jeremy’s reading partner isn’t a tool-it’s an agent requiring context, capable of exploration, and collaborative in nature.

• Context Engineering as Literacy: Just as information literacy became crucial in the internet age, context engineering is becoming essential in the AI age. Knowing how to prepare and manage context for LLMs is a new form of literacy.

• The Evolution of Reading: From oral tradition to written text to printing press to digital text to AI-assisted reading-each transition changed how humans engage with knowledge. This video captures the beginning of the next evolution.

• Critical Thinking in the AI Age: Rather than replacing critical thinking, LLMs can augment it by making verification and exploration easier. Jeremy’s skeptical approach shows how AI can support, rather than supplant, rigorous thinking.

• The Democratization of Expertise: Just as Google democratized information retrieval, LLMs are democratizing expertise. Complex books previously requiring extensive background knowledge can now be engaged deeply by general readers.

• AI-Augmented Learning: This demonstrates a template for how AI can transform education-not by replacing human engagement but by amplifying it, enabling deeper exploration and personalization.

🛠️ FRAMEWORKS & MODELS

The Context Engineering Framework

A systematic approach to preparing an LLM for deep engagement with complex material.

• Components:

  • Hierarchical summaries: Chapter, part, and whole book summaries for multi-level understanding
  • Personal context: Your background, situation, and relevant prior discussions
  • Source materials: Footnotes, references, and supporting documents accessible to the LLM
  • Reading objectives: What you want to get out of the book and specific questions to explore

• How it Works: Instead of asking an LLM to “read this book,” you provide it with a carefully curated information hierarchy that fits within context windows while maximizing comprehension. The LLM can then draw connections between the current text and the broader context you’ve provided.

• Significance: This framework explains why most people get mediocre results from LLMs-they don’t invest in context engineering. The two hours of preparation Jeremy describes transforms the LLM from a generic assistant into a specialized reading partner.

• Evidence: Jeremy’s demonstration shows the complete setup process and its results. Eric’s surprise at the depth of preparation validates that this goes beyond typical usage patterns.

The Curiosity-Driven Exploration Model

An approach to reading that treats rabbit holes and tangents as valuable rather than distractions.

• Components:

  • Follow interest: When something catches your attention, explore it deeply
  • LLM as research assistant: Use the AI to gather context, verify claims, and find connections
  • Generous interpretation: Assume there’s more nuance than appears on the surface
  • Contemporary context: Connect historical events to present-day implications

• How it Works: Traditional reading, especially of business books, often focuses on extracting main ideas efficiently. This model inverts that-following curiosity wherever it leads, using the LLM to explore historical context, verify claims, and understand connections. The assumption is that the value lies in the exploration, not just the conclusion.

• Significance: This model is only viable with LLM assistance because the exploration happens at machine speed. Without AI, following every interesting tangent would be impossibly time-consuming. With AI, it becomes a rich, engaging experience.

• Evidence: Jeremy’s exploration of the Boeing/GE management lineage, the Bentham/Thompson contrast, and the Sol Price/Amazon connection all demonstrate this model in action. Each started with curiosity and led to deeper understanding.

The Skeptical Verification Protocol

A systematic approach to maintaining critical thinking while reading with AI assistance.

• Components:

  • Fact-checking: Verify specific figures and claims by asking the LLM to check sources
  • Editorial scrutiny: Question whether statements are facts or interpretations
  • Counterexample seeking: Ask for cases that might contradict the author’s thesis
  • Clarification requests: When something seems odd or unclear, dig deeper

• How it Works: Rather than accepting the text (or the LLM’s interpretation) at face value, this protocol treats both as hypotheses to be tested. The LLM becomes a partner in verification-checking footnotes, finding counterexamples, clarifying ambiguous passages.

• Significance: Many worry that AI will make people less critical. This protocol shows the opposite-AI can enable more rigorous verification by making it faster and easier to check claims and find counterevidence.

• Evidence: Jeremy repeatedly fact-checks figures, questions whether something is “editorializing,” and asks the LLM to verify claims. The Boeing discussion includes specific verification of the Welch management lineage and its consequences.

The Session Continuity System

A method for maintaining context and continuity across multiple reading sessions with an LLM.

• Components:

  • Handoff notes: Detailed explanations written by the LLM for “future me” (the next session)
  • Incremental accumulation: Each session adds to the accumulated handoff notes
  • Quick setup: Template-based approach to starting new chapters (duplicate, paste, adjust)

• How it Works: Since LLMs have limited context and each session starts fresh, this system creates a written record of everything the “new” LLM needs to know. Jeremy’s “Good luck future me” notes include summaries of previous reading, key insights, his personality and preferences, and context about his situation.

• Significance: This solves the practical problem of session boundaries in LLM usage. More importantly, the act of creating handoff notes forces synthesis and reflection, improving retention and understanding.

• Evidence: Jeremy shows the actual handoff notes and his process for creating them. He estimates it takes “less than five minutes now, like maybe two or three minutes to start each new chapter.”

💬 NOTABLE QUOTES

1. “It’s like talking to somebody who knows everything about my experience and has read and it’s it’s got the whole of the big uh the previous chapter, the next chapter, it’s got the part summary, it’s got the whole book summary.” Context: Jeremy describing the LLM’s comprehensive context Significance: Captures the transformative nature of context engineering-the LLM becomes a reading partner with comprehensive background knowledge that would be impossible for a human assistant.

2. “I wouldn’t have put into preparatory work here instead what I would have done is done something like far easier gotten a halfway a good result, then tried to use it in context and gotten I would have gotten stuck downstream of this and then be like, ‘Oh, the LM’s not good at this. Forget it.’” Context: Eric admitting he would have skipped the deep context preparation Significance: Highlights why most people get poor results from LLMs-they don’t invest in context engineering. Eric’s surprise validates that Jeremy’s approach goes far beyond typical usage.

3. “Compared to time you’re going to spend to read a book, it’s like two hours to read like to make your reading of a book, you know, dramatically more effective is such a good investment. And yet even even now I’m like two hours that is sounds like kind of a lot, you know.” Context: Jeremy on the context engineering time investment Significance: Acknowledges the psychological barrier to preparation even while arguing for its value. The “two hours” becomes a memorable benchmark for the investment required.

4. “This is kind of a new primitive um that’s different. I think it it’s probably something that just technologically wasn’t possible until a small number of months ago.” Context: Jeremy on the novelty of this reading approach Significance: Positions this as a genuine paradigm shift rather than an incremental improvement-the capability literally didn’t exist before recent LLM advances.

5. “Good luck future me.” Context: The LLM’s sign-off in the handoff notes Significance: Charming and memorable demonstration of treating the LLM as a collaborative partner across time. The “future me” framing makes session continuity concrete and personal.

6. “I’m a big fan of this guy called Peter Wozniak who created this incremental reading approach… it’s very good at this literal kind of glittery analysis, don’t you think, Eric?” Context: Jeremy connecting his approach to existing learning methodologies Significance: Shows how LLM-assisted reading builds on established learning science (incremental reading, spaced repetition) while adding new capabilities.

7. “You are both you are like a highly skeptical but also good faith reader which I find to be a very um unusual combination.” Context: Eric describing Jeremy’s reading approach Significance: Identifies the ideal stance for AI-assisted reading-skeptical enough to question, good-faith enough to be convinced. This balance is crucial.

8. “I don’t know like this is one of the absolute best reading experiences I’ve ever had. Like I’m I’m trying to think is it the best?” Context: Jeremy reflecting on the experience Significance: Powerful endorsement from someone with decades of reading experience. The comparison to reading Dawkins and Wolfram establishes this as a peak learning experience.

⚡ APPLICATIONS & HABITS

Practical Guidance

For Readers:

• Invest 2 hours in context engineering before starting a complex book. Create hierarchical summaries, gather personal context, and organize reference materials.
• Use the LLM as a skeptical partner-regularly ask “is this true?” and “what am I missing?”
• Follow rabbit holes that interest you. The LLM makes exploration efficient enough to be worthwhile.
• Apply concepts to your own situation by sharing your context with the LLM.
• Maintain continuity with handoff notes between sessions.

For Authors:

• Recognize that AI-assisted readers will engage with your work at unprecedented depth.
• Expect skeptical verification-claims will be checked, references explored.
• Consider how allusions and references will land with readers who can instantly look up context.
• Appreciate that readers can follow connections you only hint at.

For Educators:

• Teach context engineering as a new form of literacy.
• Demonstrate skeptical verification as a standard practice.
• Encourage curiosity-driven exploration over efficient information extraction.
• Show students how to maintain continuity across learning sessions.

Implementation Strategies

Week 1: Setup

• Choose a complex book you want to read deeply
• Create hierarchical summaries (chapter → part → whole)
• Gather personal context relevant to the book’s topics
• Organize footnotes and references for easy access

Week 2-4: Deep Reading

• Read with the LLM as a partner
• Follow rabbit holes that interest you
• Verify claims that seem questionable
• Apply concepts to your own situation
• Create handoff notes at the end of each session

Ongoing: Refinement

• Develop your own prompting patterns
• Build templates for different types of books
• Share your approach with others
• Iterate based on what works for you

Common Pitfalls to Avoid

Insufficient Context

• Don’t expect good results without investing in context engineering
• The two hours of setup is crucial-skipping it leads to superficial engagement
• Hierarchical summaries are essential for multi-level understanding

Passive Consumption

• Don’t treat the LLM as a summarizer-treat it as a partner
• Passive acceptance of both text and LLM output defeats the purpose
• The goal is dialogue, not delegation

Skipping Verification

• Don’t accept claims at face value-use the LLM to verify
• Fact-check figures and specific assertions
• Question whether something is fact or interpretation

Ignoring Session Boundaries

• Don’t start each session from scratch-maintain continuity
• Handoff notes are essential for preserving context
• Without continuity, each session feels like starting over

Forgetting the Human Element

• The LLM assists but doesn’t replace your judgment
• Your curiosity, skepticism, and interests drive the exploration
• The LLM amplifies your engagement-it doesn’t replace it

📚 REFERENCES & SOURCES CITED

• fast.ai: Jeremy Howard’s AI education platform, demonstrating his expertise in practical AI application.

• The Lean Startup: Eric Ries’s previous bestselling book, establishing his credibility as an author and thought leader.

• Sol Price / FedMart: Founder of FedMart, whose retail philosophy influenced Costco and Amazon. Chapter 1 of Eric’s book focuses on his story.

• Jeremy Bentham: English philosopher mentioned in the book, known for utilitarianism.

• Jack Welch / GE: Former CEO of General Electric, whose management approach influenced multiple executives who later led Boeing.

• SuperMemo / Incremental Reading: Peter Wozniak’s spaced repetition and incremental reading methodology, which Jeremy references as an influence.

• “How to Read a Book”: Reference to Mortimer Adler’s classic guide to reading, mentioned in the context of close-reading traditions.

• Andy Matuschak: Designer and researcher mentioned in the conversation, known for his work on tools for thought.

• Answer AI: Jeremy Howard’s current company, the context for his personalized application of the book’s concepts.

• Boeing / 737 MAX: Referenced as a case study in the book and explored in depth during the demonstration.

⚠️ QUALITY & TRUSTWORTHINESS NOTES

• Accuracy Check: Jeremy’s claims about his process are demonstrated in real-time, making them verifiable. The specific figures he fact-checks (e.g., Welch’s GE performance) are confirmed through the footnote system he set up.

• Bias Assessment: Both Jeremy and Eric have clear enthusiasm for this approach, but they’re transparent about its limitations and the learning curve. Jeremy acknowledges the two-hour setup time might seem like “a lot.” No commercial stake in promoting specific tools beyond genuine preference.

• Source Credibility: Primary source-this is Jeremy demonstrating his actual workflow. Highly credible for his personal experience. Eric’s commentary adds authorial validation of how the reading experience extracts value from his work.

• Transparency: Excellent transparency about the experimental nature of the approach, the time investment required, and what might not work. Jeremy openly shares his skepticism and verification process.

• Potential Harm: Low risk of harm. The methodology requires critical thinking and verification, which are protective factors. The main risk is that readers might expect instant results without the context engineering investment, but this is addressed explicitly in the video.

🎯 AUDIENCE & RECOMMENDATION

Who Should Watch:

• Avid Readers: Essential viewing if you read complex books and want to engage with them more deeply. This transforms what’s possible.

• Students and Researchers: Critical for understanding how AI can augment deep learning and research. The methodology applies broadly beyond books.

• Authors: Highly relevant for understanding how AI-assisted readers will engage with your work. May influence how you write.

• Educators: Important for understanding how learning is evolving and how to teach AI-augmented literacy.

• Knowledge Workers: Useful for anyone who needs to engage deeply with complex written material as part of their work.

Who Should Skip:

• Casual Readers: If you only read for entertainment or prefer not to engage deeply with texts, this methodology will feel like overkill.

• Those Without LLM Access: The approach requires access to a capable LLM (Claude, GPT-4, etc.). Without that, the video is theoretical rather than actionable.

• Those Seeking Entertainment: This is a technical demonstration, not a polished documentary. The value is in the information density and specificity.

Optimal Viewing Strategy:

• Speed: 1.25x-1.5x speed works well for the demonstration portions. The conversational sections with Eric are worth listening at normal speed.

• Sections to prioritize:

  • 00:00-07:00 - Context engineering setup (essential for understanding the methodology)
  • 12:00-25:00 - Reading demonstration with rabbit holes (shows the approach in action)
  • 35:00-39:00 - Session continuity and handoff notes (practical implementation details)
  • 42:30-43:30 - Meta-reflection with Eric (philosophical implications)

• Note-taking: Take notes on the specific workflow, prompts, and context preparation steps. These are actionable and specific.

• Follow-up: After watching, try implementing the context engineering approach with a book you’re planning to read. The two-hour investment will make the methodology concrete.

Meta Notes: This video rewards careful viewing and note-taking. It’s information-dense and demonstrates rather than explains. The conversation with Eric adds crucial validation and meta-commentary. This is likely to become a reference video for AI-assisted learning methodologies.

Crepi il lupo! 🐺