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 30 days

longest streak 4 days

675 / 1,000 XP325 XP to wheel

Sat, Oct 31

If training a model on its own outputs leads to model collapse, what happens when you continually give it it's own code to build on?
Im not a big expert in this but it seems that something similar happens with the code. It is slowly degrades I think. Short iteration loops, diff reviews, static analysis and etc will steer it.

Mon, Oct 26

Thu, Oct 22

Tue, Oct 20

So since dev/null is written with the Laravel + Inertia stack, I'll use it as an example of a problem I'm curious how ya'll are solving. When I click on a navigation link on the left, there's a latency between clicking where there are no UI updates until the round trip is…
I feel this is where html-on-the-wire loose against traditional spa / back pattern. For a X-like app with a lot of navigation, react or SolidJS with dedicated separated api backend is more work but offers nicer UX.

Mon, Oct 19

I'm curious, if Cloud LLM vendors (Anthropic, OpenAI, Google, etc) decided to increase the price of AI access by 10x, would you keep paying? E.g. $20/mo -> $200/mo or $200/mo -> $2,000? If so, why?
  • Fork out the cash0%, 0 votes
  • Back to coding by hand17%, 1 vote
  • Switch to local LLMs (including hardware costs)83%, 5 votes
  • Something else (comment below)0%, 0 votes

6 votesFinal results

That's @AidanTilgner's root commit.