I have found putting the spec together with a model, having it to try find blindspots and write done the final take in clear and concise language, useful.
A good next step is to have the model provide a detailed step by step plan to implement the spec.
Both steps are best done with a strong planning model like Claude Opus or ChatGPT5, having it write "for my developer", before switching to something like Claude Code.
I have found Claude code to be significantly better, both in how good the model ends up being and in how polished it is. To the point that I do not drop down to Gemini CLI when I reach my Claude usage limit.
The first step is to acquire hardware fast enough to run one query quickly (and yes, for some model size you are looking at sharding the model and distributed runs). The next one is to batch request, improving GPU use significantly.
Take a look at vLLM for an open source solution that is pretty close to the state of the art as far as handling many user queries:https://docs.vllm.ai/en/stable/
One thing I could not find on a cursory read is how used were those developers to AI tools. I would expect someone using those regularly to benefit while someone who only played with them a couple of time would likely be slowed down as they deal with the friction of learning to be productive with the tool.
In this case though you still wouldn't necessarily know if the AI tools had a positive causal effect. For example, I practically live in Emacs. Take that away and no doubt I would be immensely less effective. That Emacs improves my productivity and without it I am much worse in no way implies that Emacs is better than the alternatives.
I feel like a proper study for this would involve following multiple developers over time, tracking how their contribution patterns and social standing changes. For example, take three cohorts of relatively new developers: instruct one to go all in on agentic development, one to freely use AI tools, and one prohibited from AI tools. Then teach these developers open source (like a course off of this book: https://pragprog.com/titles/a-vbopens/forge-your-future-with...) and have them work for a year to become part of a project of their choosing. Then in the end, track a number of metrics such as leadership position in community, coding/non-coding contributions, emotional connection to project, social connections made with community, knowledge of code base, etc.
Personally, my prior probability is that the no-ai group would likely still be ahead overall.
FWIW, LLM tooling for Emacs is great. gptel for example allows you to converse with wide-range of different models from anywhere in Emacs — you can spontaneously send requests while typing some text or even browsing M-x menu. I often do things like "summarize current paragraph in pdf document" or "create a few anki cards based on this web page content", etc.
Yes! I recently had to manually answer and close a Github issue telling me I might have pushed an API key to github.
No, "API_KEY=put-your-key-here;" is a placeholder and I should not have to waste time writing that.
I don't use it to avoid reading man pages. Rather, as often with LLMs, this is a faster way to do things I already know how to do. Looking at commands I run in various situations and typing them for me, faster than I can remember the name of a flag i use weekly with a pdf processing tool or type 5 consecutive shell commands.
Money wise, my full usage so far (including running purposely large inputs/outputs to stress test it) has cost me.... 19c. And I am not even using the cheapest model available. But, you could also run it with a local model.
Yes, it is API based and uses your last unique 100 shell commands as part of its prompt: it seemed important to remind users that this data does leave their machine.
A fork using a local model should be fairly easy to set up.
I think the top post on the Krita thread does a pretty good job at setting their boundaries. Something that cannot replace artists: it will not "beautify" art, and stays close to the input, also it should not be trained on the work of unwilling artists.
Anthropics actually encourages using Claude to refine your prompts! I am not necessarily a fan because it has a bend towards longer prompts... which, I don't know if it is a coincidence that the Claude system promps are on the longer side.
It doesn't merely encourage, for at least a year now, they've been offering a tool for constructing, improving and iterating on prompts right in their console/playground/docs page! They're literally the "use LLM to make a better prompt for the LLM" folks!
For people who want a non web-based alternative, these days I use Xournal++ (https://xournalpp.github.io/) to do that type of edition locally.
What I am still looking for is a good way to clean scanned PDFs: split double pages, clean up text and make sure it is in lack and white, clean deformations and margin, cut and maybe rotate pages, compress the end result.
For the other issues, I haven’t found any single good tool, but I’ve stitched together things like unpaper, ghostscript and deskew ( https://github.com/galfar/deskew ).
Also, if you need OCR, hocr-tools and Google’s Document AI ocr API have worked really well for me (I tried Gemini, but you run into issues with big documents).
A good next step is to have the model provide a detailed step by step plan to implement the spec.
Both steps are best done with a strong planning model like Claude Opus or ChatGPT5, having it write "for my developer", before switching to something like Claude Code.
reply