AI is just a tool

If you used a fancy auto bake cake machine instead of an oven, you still get to claim that you made the cake.

100 years ago someone would be making the claim that using an oven to make cakes “doesn’t count”

All AI did was raise the bar

It’s quite clear here that the author spent a lot of time on this so he absolutely gets credit as the author

Do you think karpathy's autoresearch would be useful here?

OpenAI and others are already profitable on inference (inference is really really cheap)

They are just heavily investing into the latest frontier

The biggest risk is whether they can stay cutting edge, or if open source or others will catch up quickly.

Your goal.md examples are all features for the existing codebase. Any largish goal.md examples where your system is able to 1 shot a pretty large app?

The goal.md is what makes this thing either amazing or terrible for the user, so any guidelines or clear examples on writing a good one would go a long way.

Could you push it even further? Infini video zoom?

I keep zooming and its just videos all the way down

https://labs.ramp.com/rct

"claude code plays wall street raider" would be very very cool.

https://www.anthropic.com/engineering/building-c-compiler

Here is the relevant excerpt:

"To prevent two agents from trying to solve the same problem at the same time, the harness uses a simple synchronization algorithm:

Claude takes a "lock" on a task by writing a text file to current_tasks/ (e.g., one agent might lock current_tasks/parse_if_statement.txt, while another locks current_tasks/codegen_function_definition.txt). If two agents try to claim the same task, git's synchronization forces the second agent to pick a different one. Claude works on the task, then pulls from upstream, merges changes from other agents, pushes its changes, and removes the lock. Merge conflicts are frequent, but Claude is smart enough to figure that out."