> Our hypothesis is simple: session logs are now the most important artifact in software development, and should be stored alongside the code itself in the repository.
Pi.dev has a feature where you can export the session as a html file and look at it later. I foresee that potentially you could store this in the same Git repository and get the benefit of reviewing how a particular code change came about during a session with an agent.
I guess the next step would be having the coding agent save that session context automatically in a folder in the git repository rather than requiring a human to export it.
This startup also seems to be operating in a similar space to tangled.org - moving code repos into a decentralised hosting environment.
- Unit tests
- End to End testing with Cucumber and Puppeteer (or Playwright these days)
- Code Coverage analysis
- Lint and formatting checks
- Tools like knit to remove unused code
- Security checking tools like Synk and Socket
- Size checking
If you look at what those tools do, they automate aspects of reviewing PRs. The question then becomes - what still gets done manually, and are there ways to automate that.
Meditation is great, but there might also be other options available when it comes to dealing with the effects of context switching.
For me, I find the Pomodoro technique really effective, but there's a key caveat, which is that you try to focus on one thing at a time (which flies in the face of using multiple AI agents working on x number of things at the same time).
Another angle to explore is how much of the process of software development that we do manually can we automate, particularly the parts that still require human input (like code reviews). That may also help with reducing the cognitive load.
Recently I got AI to help make a script to convert a CSV dump of trading activity on a trading platform for uploading into another stock trading analysis platform.
Later on, I managed to crunch the numbers in the script, and realised I could dump them out and display them in a dashboard alongside the trade activity. So I built that too,
An excel spreadsheet could have done the same job as the dashboard, but the script for conversion was greatly aided by the AI tool. The work otherwise would have been a bit of manual coding and back-and-forth testing.
I find it nuts that I can use Claude Code for $20pm - I imagine that won't last forever but have to say it is great value for money.
So when I see monthly budgets in the thousands for developers at some larger companies, I'm curious to learn how they are managing to spend that kind of figure: how much code/documentation are they feeding into their prompts, are they using agent orchestration systems to make the code factory run 24/7, and how much value is coming out the other end versus before?
And, if they are pouring thousands into LLMs per developer, have they considered looking at alternatives like having LLMs running locally on own hardware with their own agent harness?
Those are the kind of questions I'd love to ask - I just wonder how much stuff is truly cutting edge and how much might be wasteful?
I would say the first time I did “vibe coding”, when I tried Claude Code with Zed’s agent integration in January this year.
I wanted to see if I could build an image editor for isometric graphics using HTML5 canvas, Svelte, Vite, and the. Rather than do all of the skeleton code setup, I figured “why not try and see if Claude can build the app scaffolding?”.
I gave it a prompt and watched it produce the scaffold, along with a few features I outlined in the prompt.
When I booted the app and saw that the features worked and that there had been an element of design to the layout, that was my mind-blown moment. In a period of about 45 minutes, I added some features and had a basic MVP at the end. I walked back home stunned.
The Financial Times' Unhedged Podcast covered the SpaceX IPO recently and highlighted the same issues that the Danish pension fund raised concerns about.
I've been able to get by with the $20pm Pro subscription and reap great value out of Claude Code.
I feel like it really is about:
- Don't feed it the works of Shakespeare into the context window if all it's working on is a few files. I actually don't have a Claude.md file in my projects.
- I write the prompt as if I was giving instructions to another developer or to myself on how I want to approach a specific coding, with a numbered step plan. I've actually been able to take the details written into a Jira ticket on a work project, feed it into Clade Code, and get really good results from it.
- If you are responsible for the output, then you need to review the output - that does put a natural constraint on the tool's usage, but ultimately it is you who uses the tool, not the other way around.
I feel like that's the thing - you have to find the right cadence, just like with running or driving a car - you need to find the level at which you control the car, at which you maintain a consistent pace, and at which you get code that does what you need it to do and meets the quality threshold you want.