> “One day, the successor to Claude Code will make a superhuman interface available to everyone. And if Tokens were TCP/IP, Claude Code is the first genuine website built in the age of AI. And this is going to hurt a large part of the software industry.”
“I believe that all software must leave information work as soon as possible. I believe that the future role of software will not have much ‘information processing’, i.e., analysis. Claude Code or Agent-Next will be doing the information synthesis, the GUI, and the workflow. That will be ephemeral and generated for the use at hand. Anyone should be able to access the information they want in the format they want and reference the underlying data.
What I’m trying to say is that the traditional differentiation metrics will change. Faster workflows, better UIs, and smoother integrations will all become worthless, while persistent information, a la an API, will become extremely valuable.”
Don't get me wrong, I somewhat agree that there's been a sea change with Opus 4.5 in the usefulness of Claude Code, but it stills goes off the rails at the drop of a hat in the dumbest and most frustrating ways. Actually trying to use it to develop even a nontrivial greenfield project from scratch requires carefully reviewing its code to make sure it stays on track.
Completely agreed!
We're less than a calendar year away from AGI and ignorant luddites still want to pretend the entire world hasn't shifted under their feet.
Genuinely confused here.
Note: Figma was just an example of a complex software.
Why even generate an image then. Why don I just use dribbble / behance and pick the best design
Because I have neither the time nor inclination to make it at home right now. I have other stuff I need to do.
Even hardcore top level engineer wont be able to do it just via LLMs
What closed source software company can compete on features against the entire world's collective agents? What sales pitch do hosted saas vendors have when you can spin up a container for your open source saas of choice pre-configured, and the agent can tweak stuff and offload ops efficiently? We're ~2 years away from this reality.
Most of the recent completely vibe coded projects dont do anything as complex as that, they are usually just tiny websites or some slop tui
I have used all major tools: OpenAI (chat, api), Google Gemini (ai studio, api, cli, antigravity) and Claude (chat, code and api). Mostly for coding issues to solve.
Claude Code gives usable results almost instantly for small scripts and it can go live. Gemini CLI tells me that it doesn't have this and that - and I have tried pushing Gemini to deliver production-quality code. No chance.
I use the same style of coding instructions for all tools.
But difference is in hours. I had a Claude session - result was in minutes, Gemini - hour and in many rounds.
On the other side, Gemini Canvas is really powerful as it makes usable app/tool inside Gemini so you don’t have to know how to run Python or PHP.
And OpenAI has very powerful chat.
So all of them seem to have different focus groups…
But for your use-case, I would think Gemini cli is better, since gemini does seem to excel in Data Analytics tasks for me.
I would like to direct the author to a high school statistics class.
https://en.wikipedia.org/wiki/Correlation_does_not_imply_cau...