avandekleut

Solved theoretically? I figure you could at least intuit that reversing all the moves in the scramble in reverse order would solve it.

Are you me? We do the same thing, and they also translate it into hours by averaging across the whole team's velocity.

Tailwind is a UI styling and components company. Are you thinking of Tailscale?

As a technical founder I empathize with the authors situation and approach, but honestly spending that much time and energy on a product before even getting one sign up is a known failure mode.

not to mention that dopamine is generally associated with anticipation and searching + reinforcing behaviours, whereas pleasure and satisfaction is associated with the opiate system

I remember spending hours and hours on pivot and forums like droidz.org which used to host animations, models, and forums. I even remember learning about "easing" and different "levels" of animation and collabs as showcased on darkdemon.org[1]

[1] https://www.youtube.com/watch?v=vkqDoAYKG4A

It takes something like 60 turns or so in my experience. you probably messed up while doing it.

Using context-free grammars and sampling techniques.

Please consider making this responsive on mobile!

Also want to shout out ts-rest. We have a typescript monorepo where the backend and frontend import the api contract from a shared package, making frontend integration both type-safe and dead simple.

Ive found similar behaviour with stopping at linting errors. I wonder if my instructions are conflicting with the agent system prompt.

You may be interested in reading Manna, a short story about AI that operates exactly this way.

Someone should implement this using tool calls.

Ive used smithery.ai as well.

MCP also has support for "hints" which note whether an action is destructive.

Another common pattern is to put the "primary" argument first, and the rest in an "options" object argument.

Typescript nukes primitive aliases like this sadly. Intellisense will just infer it as "string".

Similarly to passing an object argument, you can return an object.

Then pair it with destructuring assignment!

`const { fullName } = getName({ firstName, lastName )}`

2:12 27 moves, my first guesses for words were correct. The Q was the key for me. Fun game!

I'm glad I'm not the only one that has found 3.5 to be better than 3.7.

I used SpinningUp as one of my main sources for my thesis. I also wrote some downloadable-and-runnable blog posts here: https://avandekleut.github.io/ppo/

Good luck! Your username suits you.

This is pretty much my flow that I landed on as well. Dump existing relevant files into context, explain what we are trying to achieve, and ask it to analyze various approaches, considerations, and ask clarifying questions. Once we both align on direction, I ask for a plan of all files to be created/modified in dependency order with descriptions of the required changes. Once we align on the plan I say lets proceed one file at a time, that way I can ensure each file builds on the previous one and I can adjust as needed.

I started with cosine law and then went "wait a minute" - but still jumped to the wrong answer as pointed out in the solutions.

That sounds like a perfect use case for my new collaborative diagramming and whiteboarding app!

I always ask it to "analyze approached to achieve X and then make a suggestion, no code" in the chat. Then a refinement step where I give feedback on the generated code. I also always try to give it an "out" between making changes and keeping it to same to stave off the bias of action.

I just finished interviewing with a company called Infisical. The take-homes were crazy (the kind of thing that normally takes a few days or a week). I was paid but it took me 12 hours.

I had to refresh the page for the layout to work correctly on mobile Safari

what did you decide to do?

Can you elaborate on some of the issues you faced? I was considering deploying to ECS fargate as we are all-in on AWS.