> I agree, but I think I'm less optimistic than you that Claude will be able to catch its own mistakes in the future. On the other hand, I can definitely see how a ~more intelligent model might be able to catch mistakes on a larger and larger scale.
Claude already does this. Yesterday i asked it why some functionality was slow, it did some research, and then came back with all the right performance numbers, how often certain code was called, and opportunities to cache results to speed up execution. It refactored the code, ran performance tests, and reported the performance improvements.
I have been reading through this thread, and my first reaction to many of the comments was "Skill issue."
Yes, it can build things that have never existed before. Yes, it can review its own code. Yes, it can do X, Y and Z.
Does it do all these things spontaneously with no structure? No, it doesn't. Are there tricks to getting it do some of these things? Yup. If you want code review, start by writing a code review "skill". Have that skill ask Opus to fork off several subagents to review different aspects, and then synthesize the reports, with issues broken down by Critical, Major and Minor. Have the skill describe all the things you want from a review.
There are, as the OP pointed out, a lot of reasons why you can't run it with no human at all. But with an experienced human nudging it? It can do a lot.
It's basically not very different from working with an average development team as a product owner/manager: you need to feed it specific requirements or it will hallucinate some requirements, bugs are expected, even with unit test and testers on the team. And yes, as a product owner you also make mistakes, never have all the requirements up front, but the nice thing working with a GenAI coder is that you can iterate over these requirement gaps, hallucinated requirements and bugs in minutes, not in days.
What works best for me using Claude Code is to let the CC engineer its own context. You need to provide it with tools that it can use to engineer its context. CC comes with a lot of tools already (grep, sed, curl, etc), but for specific domain you may want to add more, e.g., access to a database, a cms, a parser for a bespoke language, etc.
With these i'll mostly just give it questions: what are some approaches to implement x, what are the pros and cons, what libraries are available to handle x? What data would you need to create x screen, or y report? And then let it google it, or run queries on your data.
I'll have it create markdown documents or skills to persist the insights it comes back with that will be useful in the future.
LLMs are pretty good at plan/do/check/act: create a plan (maybe to run a query to see what tables you have in your database), run the query, understand the output, and then determine the next step.
Your main goal should be to enable the PDCA loop of the LLM through tools you provide.
According to Claude (easy guess from the wikipedia link?):
The book is almost certainly by *Franco Moretti*, who coined the term "distant reading." Given the timeframe ("maybe a decade ago") and the description, it's most likely one of these two:
1. *"Distant Reading"* (2013) — A collection of Moretti's essays that directly takes the concept as its title. This would fit well with "about a decade ago."
2. *"Graphs, Maps, Trees: Abstract Models for Literary History"* (2005) — His earlier and very influential work that laid out the quantitative, computational approach to literary analysis, even if it didn't use "distant reading" as prominently in the title.
Moretti, who founded the Stanford Literary Lab, was the major proponent of the idea that we should analyze literature not just through careful reading of individual canonical texts, but through large-scale computational analysis of hundreds or thousands of works—looking at trends in genre evolution, plot structures, title lengths, and other patterns that only emerge at scale.
Given that the commenter specifically remembers learning the term "distant reading" from the book, my best guess is *"Distant Reading" (2013)*, though "Graphs, Maps, Trees" is also a strong possibility if their memory of "a decade" is approximate.
As a windsurfer, wingfoiler, and kitesurfer i can only say that both Windy.app a nd Windy.com are awesome. Like the ability to compare different models easily.
How do you know the end result is worse than disinformation? If the Russian disinformation allow Russia to destroy the freedom and democracy in Europe, and allow Russia to take over, that seems to me to much worse than limiting the publication of lies and slander.
Something doesn't need to be slander to be information warfare. We need something much larger scale and more powerful to fight back the hybrid warfare from Russia and increasingly also the US.
Is it possible that that belief is itself influenced by the propaganda you’re worried about? If I were Russia I’d do my best to create a rift between the US and Europe.
Freedom is a scale, not binary. I'm willing to move a bit on that scale to avoid going completely to one of the opposites. I completely disagree with your suggestion that if you don't have complete freedom, you're at the complete opposite end, I.e. zero freedom.
Nice, the AI enabled Cambrian explosion of software is happening. Any idea can be built and validated in a few hours or days.
I've built something similar, but focused on guitar backing tracks. It's a TUI built using go, with midi output. Backing-tracks: https://github.com/ako/backing-tracks
Not the generic concept of TUI sequencers (i know, have used sequencers on c64), but any weird slightly different idea.
In my case i wanted flexible way to display lyrics, chords, strumming, metronome, fingerpicking, scales, chord shapes, etc, with an easy way for LLMs to define the backing-track, so a backing track DSL.
I think we'll see a lot of these very specialized software popping up, instead of generic solutions that contain everything and the kitchen-sink, where you don't use 90% of the functionality.
Very cool, thank you for creating and sharing. Literally yesterday YouTube searched (for the first time in a long time) for "D blues backing for Harmonica". I am interested now in exploring LLM paths for that.
You did a great job on the visualizations!!
I couldn't get sound to work on OSX yet, but will keep trying. We uke here, so will make an issue for that ;)
I work on NTCharts and also just yesterday had an LLM fix a visual bug using world understanding, one that I had thought about multiple times prior. In the end, the solution was obvious once revealed and I had overthought the problem.
Video in OP Miditui project is amazing, I missed it on first skim.
Thanks, everything was vibe coded, so visualizations thanks to vibe coding. I'm prompting like key user/product manager/architect, so expressing user needs, and ensuring the overall architecture seems reasonable.
For uke, would you expect same functionality but with 4 strings?
> Any idea can be built and validated in a few hours or days.
Let's not oversell it to that degree, please.
The amount of ideas that can be built and validated within a few hours or days went way way way up, but is still very very far away from "any". Nor will it ever get there unless we get actual AGI with "free" compute.
reply