gcampos

Thanks! We should soon have something ready to share

A lot of the points the author made about game dev with AI are things that I notice in my experience as well. That is why me and a friend started to work on https://folly.gg an AI game maker that internalizes several of these learnings.

A few months ago I had an idea of doing a no LLM hackathon for fun. The idea was to keep the coding skills sharp and have an excuse to not use Claude.

I tried that before, it doesn’t work. Claude will not prioritize AGENTS instructions the same way it did for CLAUDE.

ln was the only thing that worked for me

You cannot run Network Extensions on the simulator

Probably not directly, but indirect by the anti spam criteria: If there are too many similar apps, from whatever category, they will just reject new apps.

You could use it for long run tasks while you don’t use the laptop.

I agree with you, the whole AI thing is a distraction, even with perfect AI the whole methodology is wrong.

I like how readable the script is!

Long time ago I build my own interactive story system, but it was more focused on visuals and portability

I’m not surprised the model generate the pictures, I’m surprised that OpenAI doesn’t scan it’s own images for sexual content, violence, etc…

I have good news for you! I'm also working on a plugin device to apply VineWall protections to the whole home network.

That is something I want to eventually incorporate into VineWall.

My goal is to have VineWall to detect user patterns and use this information to help the user cope with the situations in a more healthy way

That is on my TODO list :)

That is correct! There is no annotation of which apps a packet comes from, so VineWall also runs locally a DNS proxy and uses the domain to infer the app

I graduated in Brazil almost 2 decades ago and the only places that suspended professors for being too strict were crappy for profit universities, everywhere else professors could fail the whole class if they didn’t meet standards.

Where I graduated, the hardest classes were expected to fail about half of the class

The short interview time helps keeping the interview process focused on high signal questions/discussions. That is better than a 1h where 1/3 of the process is a bunch of soft balls.

What I don’t like about them is how “dry” and mechanical the interview feels

You need to increase the context window size or the tool calling feature wont work

Ha! I think Ozempic is a better comparison :)

I have been working on a different solution (https://vinewallapp.com) for this problem, but it based on network throttling

I’m working on VineWall (https://vinewallapp.com), a network tunnel that helps you fight doomscrolling by making your internet slower when it detects you spent too much time scrolling.

At this moment I’m working on improving the logic that decides when/how much to throttle the network.

What I like it most is how "unoriginal" the result is. It is mostly the current trends, but in the future, rather than a blend of that + unknown future.

Unity was never profitable and was always dependent on vc money.

For sure it's a great product, but the truth is that it was subsidized by vc capital and is not a sustainable business

That is actually an interesting question. We track how many accidents the automated cars cause, but I'm not sure how many accidents they prevent are tracked.

To think that 1 year ago they were aggressively spamming me for an engineering position there.

How you go from "hyper growth mode" (according to the recruiter email) to closed operations within 13 months?

It's an interesting system but the reward of the bounties, specially when thinking of $/hour, makes me wonder what is the quality of the work.

I had some exploratory conversations with them last year.

The product was nice, but their business plan made zero sense to me, so I decided to not continue the process.

Technically you were right, the game is Epic ;)