memjay

This happens way too often in my experience and I tried a lot of things. If the agent has access to an existing codebase it will always try to stay as close to it as possible. It doesn’t diverge from the initial direction the project/code went into, even if it’s already clear that this direction goes nowhere. Somehow agents aren’t able to detach themselves from existing code to come up with completely new solutions.

Sometimes I feel like it would be best to only give the agent access to clear API boundaries (say public interfaces to certain modules) and let it work out a new system from scratch given the expected inputs and outputs. Then plug this independent solution into the system. Of course it can still overfit these interfaces but it’s less than having access to the entire codebase.

Thinking about it, maybe it’s possible to let one agent extract those interfaces, then use them as grounding for a new session.

One less nice way of achieving the same is to tell the agent once in a while that it should think completely from scratch (from first principles). But this relies heavily on instruction following in the reasoning part, which sometimes works and sometimes fails.

And people complain about AI water usage

Wrong url?

https://www.raycast.com/new

We have this running in production. Can get pretty expensive and slow. We are trying to replace this with cheaper and faster methods that don’t hammer our LLM and elastic search endpoints as much.

> You're still allowed to do things the old fashioned way you know.

Not if your company decides to be "AI-first" and puts agent use on your job results.

Thank you AI for explaining me how this platform works in the most unnecessarily longwinded way…

Do you give Claude access to HA somehow? Or how does it get the relevant device names/ids?

This is surprisingly delightful to play around with. We have to put some more fun into UIs, not the same type of AI-slop everywhere.

Hide this in a CLAUDE.md file of your coworker :D

> A self-hosted web app I can reach from anywhere.

So you can access the Obsidian markdown files from the web? How are you solving that?

Via VPN or do you sync your files with a server?

You might enjoy GymBook. I bought their lifetime pro version and have used it for at least 4 years now.

No ads, data can be exported, tons of exercises to choose from, live activity on the lock screen, stats, timers, measurements and much more.

Was really worth it and they seem to still be updating the app.

Edit: lifetime subscription is only ~24€

Wondering how big of a percentage have global memory across chats enabled. I always feel like those memories would sooner or later have negative impacts on output quality.

Nice write up of your findings. Enjoyed reading an article written by a real human.

Not a single screenshot on this AI generated webpage but huge buttons to pay money.

Not only hotels. I recently moved and had to endure skipping many AI-generated apartments posted to common platforms for finding new apartments.

At least some of them had an "ai generated" tag.

Agree. Systems like shadcn accelerated this moving towards sameness a lot.

Do you have a good example of a website that does not follow flat design that you like?

You are right, it feels like clickbait for blog posts / articles.

And once you notice the bait (via telltale signs) you feel betrayed. Betrayed for your attention, thoughts and time.

Do you have tips on how to get better at design, now that everything looks like the same AI-slop online?

Like how do you come up with a cohesive looking set of components for example? I use design tokens but they are not enough to get to a coherent design where components look like they belong together and kind of stand out from other designs.

Reading this hurt so much. Please just write articles by hand. I don’t care about perfect grammar and I don’t care about your article sounding „native“ or not.

But I do care about not having to read the word „genuinely“ a hundred times just because Claude likes it so much.

We decided not to go with Python/JS to make executing safe and simple.

The data structure is a recursive list of simple objects that form a table of content.

DSL uses Python syntax though. For example:

swap_section(a, b) create_section(after=2) delete_section(2)

This proved to safe a ton of explanatory prompts that would be needed if every command was a tool instead. And it’s faster and more reliable.

This has been our experience as well. Initially we had a list of tools that the agent could use to manipulate a data structure in certain ways. This approach was quite brittle. Now we are using a small DSL (domain specific language) and a single tool where the agent can input scripts written in the DSL. We are getting more dynamic use-cases now and wrong syntax can easily be catched by the parser and relayed to the agent.

> Not x — but y.

Am I the only one who feels a bit betrayed after reading LLM text? I am not even willing to try out the app after I notice… which is a shame.

At least polishing the obvious parts would help a lot and is not that much work.

Because a URL shortener is a one-line prompt today.

You will have to put a lot of effort and attention to detail to make a one-line prompt more worth than it is on its own.

Not sure whether that happened here though.

Why do all of these similar looking websites have weird broken left to right scrolling on mobile?

In addition to your experience I notice that through AI contributions our team reaches sprint goals way too early and PMs aren’t catching up with creating new tasks.

Did you notice a significant decrease in delivery speed (including debugging, fixing stuff in AI contributions)?

I wonder whether reducing AI usage might even help with cognitive load. The last few months I felt increasingly exhausted after pushing through PR after PR (although it is quite fun to do this on personal projects).

Have the same problem. Can’t see any “explore” section in the layers on iOS.

Software dev, curious about freelancing here. Just recently deleted my LinkedIn profile out of being disappointed with the value it brought to me over several years of using it. Instead, I now note down relevant contacts in a private “CRM”.

Do you think it is important to have a public facing profile of yourself and your work history for attracting new clients or is a broad in person network more relevant?