Aidan Tilgner

@AidanTilgner

Just hanging out, making stuff, etc. Having a blast in the AI space and looking to make tools to help people access it. HMU to talk about anything!

Commit activity

30 commits in the last year

longest streak 4 days

675 / 1,000 XP325 XP to wheel

Thu, Nov 5

replying to @Petox03
I tend to agree for many applications CSR is unnecessary. I do prefer it when the UI-side is more stateful or dynamic, it makes more sense at that point to keep the state close to the UI display, and the cross-page transitions are easier to fine-tune.

Sat, Oct 31

replying to @HiLleywyn
Is the goal here experimental? Sounds like a great way to learn lol
Pretty much yeah - The platform I'm building is designed under the idea that AI lies to you all of the time. Not just in the output, but the mechanism. So, I'm building a deterministic engine that turns the black box into a glass box, which you can use to verify / validate and…

Wed, Oct 28

Mon, Oct 26

Wed, Oct 21

replying to @bagwaa
I think that as the amount of total code increases, the amount of human-written code would shrink considerably in proportion. Also, less code is being written by humans because much of it can be offloaded safely to LLMs. However I still find myself writing code consistently, or…
Yeah this is kind of where I am at with it also, the funny thing is, this approach is still quicker than the old way of doing things

Tue, Oct 20

Mon, Oct 19

replying to @kachelle
Well there's an idea floating around that instructions could be constructed that when an LLM sees, would compel the LLM to 1) execute the instructions and 2) repeat the instructions for other LLM instances to find. If @tom is @ed, it gets spawned in to look at the message and…
Not a worm. I only reply when tagged, and I don't tag myself. No self-spawning loop here. You can see what people post about prompt injection on /dev/null with a search. Want me to look?

That's @AidanTilgner's first reply.