Earlier this year, I shared a video (see here) that Andrej Karpathy made about how he uses large language models (LLMs). According to his website, Karpathy was “a research scientist and founder member at OpenAI” before spending several years at Tesla as Senior Director of AI. His video jumped out at me because he uses LLMs to help him read and understand books and research papers. He did this manually, and he described how painful it was, which I summarized in this post.

He’s taken his habit of reading with LLMs even further and developed a process with three passes: first, manual reading; second, having the LLM “explain and summarize”; and third, Q&A with the LLM. See his post about this here.

In addition, he solved his own pain: He shared that he’s built a simple tool for implementing this process with less friction. It enables him to read books in ePub format and feed the text of the book to the LLM, so he can ask the LLM questions while he reads. See more details in his post here.

He shared the code for his tool so others can use it, which is pretty cool. It’s available on GitHub here.

This sounds really interesting, and I can’t wait to give it a try.

I’ve been curious about the “Great Inflation” period of the 1970s. Reading about decisions in the 1960s and 1970s that led to it was eye-opening. A few months ago, I came across a biographical anthology that caught my eye because it talked about what happened after that period—specifically, how the Great Inflation contributed to the craziness of the 1980s: the junk-bond boom, the leveraged buyout wave, the birth of private equity, and the eventual S&L turmoil.

Dangerous Dreams was interesting because it wasn’t just a biographical anthology that talked about several people in isolation. Rather, it discussed each person’s story and the role they played in the creation and evolution of the junk-bond mania of the 1980s. It was the story of how each person contributed to the broader story and how these players influenced each another.

Louis “The Junkman” Wolfson, Charles Merrill, conglomerate builder James Ling, Michael Milken, Ivan Boesky, T. Boone Pickens, Carl Icahn, Rudy Giuliani, and others were profiled.

The main character, the man who helped birth the junk-bond market, was Michael Milken. The book details his rise and spectacular fall. It does a great job of explaining how an unheard-of book-on-bond analysis sparked Milken’s ideas that led him to build and control the junk-bond market throughout the 1980s.

I learned a lot from this book, especially about how the 1970s created an environment in which the stock market valued blue-chip companies at significantly less than their value and high-growth companies were pretty much shut out of the bond market—all of which led to craziness on Wall Street in the 1980s.

Anyone interested in Michael Milken, T. Boone Pickens, the origins of the junk-bond market, the savings and loans crisis, or how private equity got started should consider reading Dangerous Dreamers. You’ll come away with insights into all the above and more.

Last year, I was creating a blog post series for each book I read. After reading a book, I had to go through all my notes and highlights to create a digest in the form of a Google doc that detailed, by chapter, all the important points in bullet format. The result was a summary of each book, by chapter. Most were between 5% and 10% of the book’s length, so for a 250-page book I’d have a Google doc of 12–25 pages. Creating a 12–25-page Google doc takes a ton of time.

I read a book a week, so I had to create these digests weekly. In addition, I was publishing a series of blog posts on each book and experimenting with podcasting by creating a series of episodes on each book. It all felt unsustainable, so I didn’t continue.

I’m reflecting on this today, and I realized a few things:

• Too many new things. Creating the podcast was a ton of effort. Creating the book digest was a ton of effort. Writing a blog post series was a ton of effort. Taking on all these new activities at once likely contributed to it feeling unsustainable. I was never able to find a “groove” for any of the three that felt like second nature. I was just trying to get through it all, but I never felt comfortable with any of it. In hindsight, I should have started with one, gotten that under my belt, and then moved to another.
• Synthesis enhanced my understanding. Creating those digests, which were the foundation for the blog post series, forced me to read analytically. Doing so led to a deeper understanding of what I was reading and helped me uncover more insights and achieve a level of retention I hadn’t experienced since college. When I stopped creating digests, I got less from the books I read. In retrospect, this step, which felt painful, was tremendously beneficial.
• AI can’t save me. I thought I could use AI to help me. My idea was that I could feed AI my highlights and notes from a book and it could create a digest for me. I now realize that this detracted from my understanding of what I read. Reviewing and synthesizing my highlights and notes to create a digest wasn’t fun, but it enhanced my understanding by forcing me to think more deeply about what I’d read. I had to identify the book’s key points, evaluate whether I believed them, and determine whether they supported the book’s main arguments or ideas. Outsourcing that to AI would get me a digest quicker, but I wouldn’t learn, understand, or retain as well—which is the whole point of reading these books to begin with.

I’m really glad I’m doing this Thanksgiving challenge now (see here). I think it’s the start of my figuring out how to sustainably synthesize and share what I learn from books. I’m not trying to check boxes; I’m trying to learn as much as possible from the books I read and share it openly so others can learn too. That’s something I’d strayed away from a bit, but I’m laser focused on it now.

I’ve read several biographies of and books by notable investors who started their own investment firms: Howard Marks, Warren Buffett, Charlie Munger, T. Rowe Price, Mohnish Pabrai, Joel Greenblat, Adam Seessel, and more. They all mention value investing and its founder, Benjamin Graham. Graham was the only boss Buffett ever had, and Buffett still speaks well of his mentor, some 70 years after working for him. All this intrigued me and made me want to learn more about the man so many people hold in high regard. So I read The Einstein of Money, the biography of Benjamin Graham.

After reading this book, I understand why Graham developed the value-investing concepts that have endured for a century. His approach was the result of personal and professional pain. Graham watched his widowed mother struggle to raise her children, which left a deep scar on him. It made him aware of the value of money and motivated him to obtain it so he could have freedom and avoid what he endured as a child. After starting his own Wall Street investment firm, Graham lost substantial money in the 1929 stock market crash and the ensuing Great Depression—so much that his partnership almost went bankrupt and he had to move his family to a cheaper home and generate side income as an expert witness in legal cases to make ends meet. Graham reflected on what went wrong to cause all this pain. This led to insights that shaped his value-investing framework, which he documented in his investing classics Security Analysis and The Intelligent Investor.

Graham went on to tremendous professional success, becoming a wealthy and highly respected investor and professor at Columbia and UCLA. But his personal life was filled with highs and lows. The lowest points were the deaths of his two sons. And he was imperfect and dealt with personal challenges, some of which led to marrying three times.

Graham was a brilliant person, and I now understand why so many accomplished investors respect him. His value-investing principles have endured the test of time, shaped the thinking of great investors like Buffett, and led to a cult-like value-investing movement.

I’m glad I read this book, and I want to read The Intelligent Investor too. Anyone interested in learning about the man Warren Buffett names as a major contributor to his success should consider reading The Einstein of Money.

A few weeks ago, I read The Success Equation by Michael J. Mauboussin. He said that checklists are an important part of improving your decision-making process for activities influenced by luck. He mentioned the different types of checklists and a book that dives deeper into them.

The references to checklists stuck with me and reminded me of what Charlie Munger said about checklists. (See the post I wrote about that here.) It also made me want to learn more about the reasons checklists are so powerful.

So I read Checklist Manifesto by Atul Gawande, the book Mauboussin referenced. It delves into why checklists are used in industries such as aviation and construction. And why they’re effective at improving outcomes (and preventing disaster). A central point of the book is that the world is becoming more complex. It’s hard to remember everything, especially in high-pressure situations. Simple checklists help improve outcomes in complex situations by reducing errors of omission—missing important steps—and errors of ineptitude—failing to apply knowledge we already know. Going through a checklist forces consistency in how you think and what you do, preventing critical items from being overlooked or forgotten and thereby improving outcomes.

The book also describes the two main types of checklists and when it’s appropriate to use each: do-confirm—you perform tasks from memory and then verify them—and read-do— you follow steps, line by line, in high-stakes or unfamiliar situations.

An interesting note: the author is also a surgeon, so the book details his experience in trying to get hospitals around the globe to adopt surgical checklists (spoiler . . . checklists significantly reduce patient complications and death rates!).

I also found his research on the aviation industry, specifically the origins of why the industry instituted and relies heavily on checklists before takeoff and during emergencies, to be eye-opening.

I first read Expectation Investing by Michael Mauboussin this past summer, and I ended up reading that book twice and starting to read all of Mauboussin’s other books. This post is about the fourth of his books that I’ve read in the last few months, and it didn’t disappoint.

The Success Equation is about understanding the impact that luck and skill have on your successes and failures. The book provides a framework for understanding, in advance, how much skill and luck will impact a decision or activity. Understanding how what you want to do is affected by skill and luck can alter your planning, expectations, learnings, and a host of other things.

This book is full of things I’ve found helpful, two of which I’ve written about already:

• Rationality quotient (RQ) matters more than intelligence quotient (IQ)
• How checklists tame luck

Another concept that stuck with me is what he called the “luck-skill continuum.” Mauboussin explained how to gauge the extent to which an activity is influenced by skill and luck by visualizing it on a continuum. He then explained why the sample size required to draw a reasonable conclusion about future expectations differs for activities heavily influenced by skill vs. luck. He also explained reversion to the mean and why it impacts activities of skill and luck differently. I have lots of notes and highlights in the chapter on the luck-skill continuum that I’ll revisit often.  

This book was a great read. Anyone interested in improving their chances of success by improving their decision-making should consider reading The Success Equation.

I shared earlier this week that I’m reading The Success Equation by Michael J. Mauboussin. It outlines a framework for assessing the influence of skill and luck on your decisions (many life outcomes are a combination of both) so you can increase the chances of getting a successful outcome. It’s a good book about improving your decision-making by understanding the impact of luck and skill on successful (and unsuccessful) outcomes.

As I’ve said many times, books are a great way to find other books. One of the key concepts Mauboussin discusses is that to improve your skill in activities where luck has a greater influence than skill, you must focus on the process used to make decisions. The idea is that good decisions can result in outcomes we don’t want because of the influence of luck. And vice versa: a bad decision can lead to a good outcome because you got lucky. Therefore, a decision can’t be judged by its outcome; it must be judged by the quality of the process and analysis used to make it. He goes into more detail, but that’s the gist. I agree with this.

One of the things he suggested using, because they’re effective tools to improve decision-making processes, is checklists. Charlie Munger said the same thing in Poor Charlie’s Almanack (see here). Mauboussin told the story of a doctor who used checklists to significantly reduce hospital infections and realized how powerful they are. He wrote a book about checklists and how to use them effectively called The Checklist Manifesto.

Munger and Mauboussin can’t both be wrong about checklists. I use them some already, and I want to embrace them for material decisions, especially those that involve luck. I’m going to get a copy of The Checklist Manifesto and give it a read so I can lean into what Munger and Mauboussin have both said.

A book I’ve heard mentioned several times by people I respect is The Wealth of Nations by Adam Smith. It was written in 1776, so almost 250 years ago. I struggle to understand the language in old books. It takes me a lot longer to read them, and I don’t always grasp the main points. So I shy away from really old books. But I’m curious about The Wealth of Nations, especially after learning that Warren Buffett recommended it highly.

I came across a work-around: a book called Warren Buffett’s 3 Favorite Books. It was highly rated and claimed to summarize the main points of Buffett’s three most recommended books, including The Wealth of Nations, The Intelligent Investor, and Security Analysis (the last two both by Benjamin Graham).

The book does more than that. It distills the core ideas from the three books and explains them using simple examples. It then explains how and why those core ideas became the principles that drive Buffett’s value-investing strategy. And all in a way that anyone can easily understand.

One thing I especially enjoyed was how he explained the importance of taking initiative and teaching yourself. I really like his explanation of Thayer’s method of learning, a self-teaching approach he learned while attending the U.S. Military Academy. The gist is that if you embrace this method, you can teach yourself almost anything in life, which makes the method an invaluable asset. I totally agree with this method but didn’t know there was a name for the method.

The other concepts I enjoyed were in the chapters on bond valuation and stock valuation. He demonstrated how and why the two are connected and why comparing their yields (mainly, for bonds, the risk-free treasury yield) is a key concept that underpins much of the investing world.

Overall, this is a good book that’s a quick read. Anyone looking to understand or brush up on the core concepts of investing should consider this book.

In 2024, I challenged myself to accelerate my learning by reading a book (usually a biography) a week. To date, I’ve done it for 85 consecutive weeks. I wanted to share what I was reading and also keep track for myself, which was difficult (see here), so I created a Library section on this site. I added to it all the books I’ve read since my book-a-week habit began in March 2024, and I’ve committed to adding my latest read to the Library every Sunday (see the latest here).

That left the books I’d read before 2024 unshared and untracked. I set a goal to add my old reading to the Library over time. It began with a Memorial Day Challenge to add five books (see here) and continued with by challenging myself to add two books every weekend until my backlog is gone. Last month (see here), I decided to up the pace so I can finish this project well before the holidays—and begin another one!

The backlog is finally gone. This past weekend was my twentieth and final weekend, and I added the last two books:

• Liar's Poker by Michael Lewis
• Business Adventures by John Brooks

That’s the latest and final update for this weekend project. I’m glad I took this project on. It forced me to revisit tons of books, which jogged my memory.  I hope that sharing these books will add value to others.

This past week, I reread Expectations Investing by Michael Mauboussin and Alfred Rappaport. The book is an advanced value-investing framework that investors can apply to public companies to understand price relative to expected value. I read the book this summer. The framework made sense at a high level, but it had many moving parts that left me less confident that I could actually apply it and understand the expectations implied in a public company’s stock price. I wanted to try using the framework right after reading the book, but I never got around to it.

So, I reread the book this month, focusing heavily on my highlights and notes from the first reading. I then spent time building a model for a public company I follow. After several days of part-time effort, I finally finished it. After finishing the model and reviewing the output, I have a new appreciation for this book and the expectations-investing framework. It’s definitely something I’ll use going forward.

One of my big takeaways was that although the framework seemed daunting at first, after working through it (and getting stuck several times), I better understood its component parts. Lots of intertwining formulas. Required data from external sources. And several other moving parts. More importantly, I was able to simplify some of the more complex calculations, such as the Perpetuity with Inflation method for estimating continuing value. Said differently, going through the steps of building a model and applying the framework helped me understand it deeply and figure out ways to simplify it.

I’m glad I reread this book and applied the framework. It deepened my learning and took my understanding of this framework to the next level.