these micro LLMs could be a game changer for hearing aids

so many interesting lowfi hardware use cases

Exactly the use cases we designed Needle for!

interesting how many resources they are putting behind this

pittsburgh has many old power plants that are perfect for data center converts

when i go home and hear about this i'm conflicted. it brings short term economic prosperity. what's the cost if there's no commitment to clean energy in reality

the developers on these projects will do whatever it takes to simply get the project done

Data centers belong in "industrial areas", ignore the snarky jokes about them being as bad as the coal plants they are replacing.

People have legitimate complaints when they are built within audible distance of housing.

Not to mention that there's plenty of room to build them well outside of developed areas altogether. Sure, there will be some logistical challenges wrt utilities, but nothing that can't be overcome - just call it a compromise.

curious if others have an elegant solution for this that gives you strong integrations + security + no provider lock in

companies like tasklet seem close for business work and there are other agent orchestration platforms like superset and conductor for coding

but wondering how others approach this problem.

i think a lazy answer is "just use claude tag". but afaik simply using claude tag doesn't make the tool calls secure

i've been thinking about this a lot. working on a longer form piece on this idea:

i find it's actually easier to "discover" than it is to "create" using llms. discovery, especially in math, is mostly a question of throwing compute at a problem and trying different results. there's a verifiable answer at the end. but there's no way (afaik) to aim something you're training via RL at a creative problem and say "make the next lord of the rings". taste literally changes by the day, and so much of what makes art matter depends on cultural significance you can't specify in advance. if you were training a music model in 1915, it would never come up with rock and roll. it's so hard to know what's going to resonate. van gogh was making his work for essentially nobody and now it's worth millions. you can't pre-load a success criteria for that

i think this is important to call out in a world that seems to devalue creative work now. people think llm generated "artifacts" are creative but in reality it's garbage and based on pattern recognition

with billion dollar seed rounds becoming the norm, it seems like there's no longer an advantage to build from within these bloated giants. can be much nimbler and have access to the same budget out the gates

i found this take interesting. i think companies like tolan and poke have made the onramp much easier for everyday agent use. i'm unsure if the cognition acq of poke is a good or bad thing (was that outcome because they couldn't find pmf?)

but i wonder who else is close to getting regular consumers to actually use "agents" rather than just a better search engine

how does this compare to superset and conductor?

There is a whole page just on this subject https://herdr.dev/compare/

Conductor is a more fully-featured app rather than a TUI (some people might prefer one or the other), and is Mac-only.