Articles on Juan Reyero

BeadHub: Coordination for AI Programming Teams

Mon, 16 Feb 2026 00:00:00 +0000

I wrote previously that the bottleneck in AI-assisted programming is shifting from individual productivity to coordination. I’ve spent the past several months building a tool to address that.

BeadHub is an open-source coordination server that lets AI programming agents claim work, talk to each other, reserve files, and escalate to humans—across machines and across programmers. I use it daily to manage around fifteen agents working on two or three products.

Beads

Around the time I wrote that article, I started using Steve Yegge’s beads, a git-native issue tracker designed for AI agents. Your agent runs bd create "Fix the login redirect bug" and it appends a JSON line to .beads/issues.jsonl, right in the repository. Issues travel with the code. When you push a branch, the issues come along.

With the Rise of AI-Assisted Programming We'll Want More Programmers, Not Fewer

Sun, 26 Oct 2025 00:00:00 +0000

TL/DR

The conventional wisdom is that AI will reduce the demand for programmers. I manage engineering teams and program with LLMs every day, and I am seeing something different.

Programmers who know how to work with LLMs can be 3-10x faster for specific tasks. Given that the size of our work backlog is only limited by our vision and ambition, these super-productive individuals will be more in demand, not less.

Data cleaning and anonymizing with GPT-3.5

Sun, 06 Aug 2023 00:00:00 +0000

Note added on 2025-07-28: It’s been two years of programming with and for LLMs. This article sounds so naïve now.

Say you made a website in which customers buy personalized gifts. Each gift comes with a message, written by the customer in whatever language the customer wants. The use of grammar, punctuation and capitalizations in the messages is often creative.

You would like to be able to offer reasonably normative messages to your customers. You would also like to store a fully anonymized version of the messages; replace all proper names with a [proper_name] placeholder, place names with [place_name], dates with [date], times with [time], and geographical coordinates with [coordinates].

The wasted talent in the meeting room

Sun, 18 Dec 2011 00:00:00 +0000

How learning about the brain could improve meetings

Go directly to the proposal for conducting meetings if you are in a hurry.

Meetings are a topic close to my heart. I have spent a large part of my professional career in them, and I must acknowledge that not all this time has been all that productive.

I started to keep an eye on other ways to merge people’s ideas ever since I realized the inescapable fact that a large part of my corporate life would be spent sitting in meeting rooms listening to people talk to each other.

Prototype your way out of uncertainty

Fri, 25 Nov 2011 00:00:00 +0000

Tinkering: prototype your way out of uncertainty

Ten years ago I found myself near Portland talking to Bill, a brilliant colleague who was working at one of HP’s printer divisions. We were both worried about the lack of tools available for the particular set of problems we were working on —figuring out the algorithms that decide which drops to print, and when, in an ink-jet printer.

It turned out we had both been thinking along the same lines: we wanted a set of libraries for image processing, written in C++ for speed, and accessible from Python for ease of prototyping. We were all avid users of the available toolkits for manipulating images (the classic pbm, ImageMagick) but there was nothing out there that could meet our rather special needs.

A graphic explanation of the Bayes Theorem

Tue, 01 Nov 2011 00:00:00 +0000

I enjoyed how the 3.16 section of the Stanford Artificial Intelligence class presented the Bayes theorem. Instead of giving a formula and expecting the alumni to apply it, they gave us a problem that the Bayes theorem would solve and expected, I believe, that we figured it out ourselves.

Being as I am counting-challenged, it took me a while to figure out a way of solving it that was simple enough that I could be reasonably sure of my results. It turned out to be a very interesting detour.

23 visits a day

Mon, 03 Oct 2011 00:00:00 +0000

Note added in 2021. When I wrote this 10 years ago I had no idea that what I then saw as a fascinating side project would become my company, and sustain my family for years. I was wondering, back then, who would want these wonderful maps I was making. The answer ended up surprisig me beyond anything I could have imagined.

The sky as seen from South Goa

I’ve been obsessed for the last 24 hours, ever since I put on-line the version of http://greaterskies.com that I thought was the first one to deserve promotion. I knew that I shouldn’t really care; I knew that, most likely, nothing would happen; I knew that I should expect to be ignored in Hacker News. And I’ve done it because it was fun. I’ve had a good time figuring out how to build the puzzle of Common Lisp, Python and Javascript that serves the star chart PDFs. And hosting is free. But still. It’s hard not to give in to what I would naturally do at this stage: add features at a frantic pace, assuming that the next thing is the one that will make the difference.

Why you should be hacking your work environment

Mon, 30 May 2011 00:00:00 +0000

I still remember my first fountain pen. It was a beautiful thing, green and blue, nice to touch, that would caress the paper when you wrote. To this day, holding a fountain pen in my hand puts me in writing mood, in thinking mood. It opens my mind, as Paul Erdös would say.

I became conscious of this not long ago when, as I was reading Daniel Kahneman’s wonderful Thinking, Fast and Slow, I understood for the first time how prevalent this effect is, and how important it probably is in my life.

Uncertainty and technology development

Tue, 18 Jan 2011 00:00:00 +0000

We cannot predict how the development of something large and complex is going to play out. We are not sure about what the right product architecture is, nor where the snags are hidden. In most cases we don’t even know what we’ll actually want by the time the product is done, when the compromises that reality will inevitably impose on us are bought off. Nor whether anybody will actually want what we envision today.

Org-mode tricks for team management

Sun, 01 Nov 2009 00:00:00 +0000

During the last three years I have managed a team of engineers doing ink-jet printer design at HP’s R&D lab in Barcelona. The excellent org-mode emacs package has helped me keep up-to-date with the work of the 12 engineers in my team, working in four different projects, without an obvious impact in my sanity.

During this time some work-flows have evolved, and I have written helper functions for org-mode to support them. This is an account of the steady state at which I have arrived. Please let me know if any of what follows could be done or written in a simpler or more idiomatic way.

A Unicycle on a Slope

Sat, 16 May 2009 00:00:00 +0000

By Juan Reyero, Jaime Fernández and Utpal Sarkar.

What is the maximum slope that a unicycle can climb? Trying to answer this question leads to some elegant results, and sheds some light on the dynamics of unicycle riding.

Assumptions

We shall assume the rotational inertia of the wheel, pedals and legs to be negligible. A further simplifying assumption is that the rider is putting all his weight into the pedal throughout its trajectory as it is going down, and that this is the only impulsion force. Which implies that the rider is not pulling up from the saddle, nor accelerating the body upwards with respect to the unicycle, nor making a horizontal force in the pedals. We know this assumption to be too stringent for expert unicyclists, who will typically be able to do all these things, but we hold it to be a reasonable approximation to what a unicyclist of normal skill would do when pressed to climb a slope at the limit of the unicycle’s possiblities.

Equilibria of a Unicycle

Sat, 16 May 2009 00:00:00 +0000

This is a very rough attempt at deriving useful conclusions from the equations of motion of a unicycle. It was inspired by my brother’s desire to buy one: my brother is large (almost 2 m tall, and around 100 Kg) and I wanted to know how the usual recommendations for beginners (a 20" unicycle) would work for him. Does a large fellow need a large wheel? The equations have turned out to be too complex of me to be able to make much sense out of them. But I haven’t lost hope.