Links

Think of me as a web crawler with taste.

How Simon Willison Uses LLMs to Help Him Write Code

Simon Willison:

Using LLMs to write code is difficult and unintuitive. It takes significant effort to figure out the sharp and soft edges of using them in this way, and there’s precious little guidance to help people figure out how best to apply them.

If someone tells you that coding with LLMs is easy they are (probably unintentionally) misleading you. They may well have stumbled on to patterns that work, but those patterns do not come naturally to everyone.

I’ve been getting great results out of LLMs for code for over two years now. Here’s my attempt at transferring some of that experience and intuition to you.

∙ AI, Tools

Life Lessons From the First Half-Century of My Career

David A. Patterson:

I started my career at Hughes Aircraft in 1972 while working on my Ph.D. at the University of California, Los Angeles (UCLA). After designing airborne computers for four years, I graduated and then taught and did systems research at UC Berkeley for the next 40. Since 2016, I’ve helped Google with hardware that accelerates artificial intelligence (AI).

At the end of my technical talks, I often share my life story and what I’ve learned from my half-century in computing. I recently was encouraged to share my reflections with a wider audience, so I’ve captured them here as 16 people-focused and career-focused life lessons.

∙ Advice, Career

A Practitioner's Guide to Wide Events

Jeremy Morrell:

Adopting Wide Event-style instrumentation has been one of the highest-leverage changes I’ve made in my engineering career. The feedback loop on all my changes tightened and debugging systems became so much easier. Systems that were scary to work on suddenly seemed a lot more manageable.

∙ Software Engineering

52 Things Kent Hendricks Learned in 2024

Kent Hendricks:

Swearing improves grip strength by 9%, wall sit time by 22%, and plank time by 12%.

This matches my experience.

The first human object launched into space wasn’t Sputnik 1. It was actually a manhole cover accidentally blown off test shaft during a nuclear test in Nevada 38 days earlier. It reached speeds equal to six times Earth’s escape velocity and was never found.

Seems appropriate.

∙ Trivia

Mark Forster's Final Version

Mark Forster:

The most distinctive feature of FV is the way that its algorithm is primarily based on psychological readiness—this then opens the way to keeping urgency and importance in the best achievable balance.

I’m a sucker for productivity systems. I like the simplicity of this one.

∙ Productivity

How to Decide After Doing Discovery

Ryan Singer:

The counterintuitive thing is, we often feel like our task is to get to a “yes.” But what we actually need is a way to say “no.” It’s the ability to eliminate many, many things that aligns us on the one thing. It’s the “no, no, no, … YES!” that gives us the power to move forward and to stick with a project.

To help us to eliminate (not forever, but for the purpose of making a decision now) I’ve found one technique very helpful. The trick is to flip things around. Instead of describing the good that will happen by doing an idea, we look at what goes wrong when we don’t do it. To make that flip, we can ask two simple questions:

  1. Knowing the customer can’t do what’s in the idea today, What are they doing instead?
  2. What’s bad about that?
∙ Product Development