Three weeks of vibe coding the new native Moolah and I’ve gone all-in. All my data has been converted over to its iCloud backend. moolah-server is still running but I haven’t been updating the data there so its getting more and more out of date. At this point, I’m all in.

AI agents like Claude Code and Codex try to be “safe” by limiting what the agent can do to a small set of operations and asking the user to approve everything else. In practice, the approval flow is so noisy and the requests so complex that it provides almost no real protection. We’ve reinvented Windows Vista’s never-ending UAC dialogs, and just like Vista, the only thing it actually trains users to do is click through.

The permission model is fundamentally broken. The fix isn’t a better dialog — it’s a sandbox.

I got Claude to crunch the numbers on the moolah-native project — eight days of vibe coding a SwiftUI replacement for my long-running personal finance tracker, with zero Swift experience. The stats are interesting but they don’t really capture what the experience has been like, so here’s the human version.

I used custom software called Moolah to track my personal finances for many years now. Originally it was written in JavaScript with moolah-server providing the backend and moolah the frontend. This holidays I’ve been vibe coding a replacement written entirely in Swift. It seems like a somewhat useful experiment to learn more about the trade offs of extreme AI usage, so I had Claude dig into the available stats from GitHub and its session logs to compare the two projects and see what we can learn. It’s not a controlled experiment so lots of room for interpretation, but an interesting data point none the less.

The initial version it came out with was a bit absurdly positive about AI, along the lines of I wrote 10x the lines of code in 0.2% of the time - more code is better! But with a bit of prompting and providing additional background it came out with the report below. I’ll write up some more human thoughts on the experience later, but I think the AI-crunched numbers are worth sharing by themselves to set the scene and because the key learnings are genuinely useful.

Claude Code has a status line that sits at the bottom of the terminal showing things like the current directory, git branch, model, and context window usage. It’s driven by a shell script that receives JSON on stdin and prints whatever it wants. I wanted to add one more thing: a short description of what the session is actually working on.

I’ve been using Claude Code a lot lately. It’s become a core part of how I work — planning changes, exploring unfamiliar codebases, writing and reviewing code. But giving an AI agent the ability to run arbitrary shell commands on your machine does make you think a bit more carefully about what’s happening on your host system.

The natural answer is to run it in a container. Not as a security boundary — Claude still needs access to your code, your git config, a GitHub token, and the internet — but as a way to keep all the side effects contained. If it installs random packages, creates temp files, or leaves build artifacts scattered around, that’s all happening inside the container rather than on your actual machine. It also makes the environment completely reproducible and disposable. Something goes wrong? Tear it down and rebuild.

So I built claude-docker to do exactly that.

The Optimism blog has published an article I wrote discussing the various types of tech debt. I’ve been finding it very useful lately to be able to “put words to things” better, whether that’s naming things better or just better ways to explain things.

Teku uses a really nice framework for separating different components - Event Channels. It’s based on similar patterns in the Sail library used at LMAX for sending network messages between services. In Teku though, it’s designed to work in-process while still decoupling the components in the system. Turns out I never wrote about it here, so I’m very belatedly catching up.

One of the downsides of moving from working on the Ethereum consensus layer is that you often need a real execution node sync’d, and they don’t have the near instantaneous checkpoint sync. So recently I bit the bullet and custom built a PC to run a whole bunch of different Ethereum chains on. I’m really quite happy with the result.

After nearly 5 years working with the ConsenSys protocols group, I’ll be finishing up at the end of January.

Mostly just so that I can find the recordings more easily later, here’s the recordings of DevCon VI talks I gave in Bogotá.

The process of producing attestations and getting them included into the chain has become more complex post-merge. Combined with a few issues with clients causing more missed attestations than normal there’s lots of people struggling to understand what’s causing those misses. So let’s dig into the process involved and how to identify where problems are occurring.