baq

16 ms·

421.
by baq 5mo ago
With the software supply chain running amok recently having anything connected feels like playing Russian roulette and I say this as somebody who is running home assistant for years. I’m particularly paranoid about connecting my ev (non-vw)
422.
by baq 5mo ago
Yeah exactly. Blowing up the rocket is the easy part. Reliably blowing up rockets on a high cadence is hard.
423.
by baq 5mo ago
On the scale of bad 1-10 where 10 is the absolutely worst case this is a 12 easily . (Elon’s strategy of blowing up smaller versions of their rockets more or less deliberately doesn’t sound so insane in the light of this.)
424.
by baq 5mo ago
None of this matters in the product: it either is capable of agentic loop workflows or it isn’t. A 10% improvement in probability of single task success makes or breaks the use case.
425.
by baq 5mo ago
> LLMs generate statistically plausible continuations of text. This works well for self-contained tasks like writing a function or drafting documentation because these are pattern‑extension problems. But pattern‑matching is not system un
426.
by baq 5mo ago
if they aren't presenting proof of doing research or they don't have the benefit of doubt (e.g. a new hire, etc.) they're being rude by not doing the research in the first place.
427.
by baq 5mo ago
efficient delivery of the single bit (and especially the flip event) to the desired audience is the use case. the actual payload almost doesn't matter as long as it's reasonably small.
428.
by baq 5mo ago
I disagree with just about everything you said being a problem except the process of cleaning up is absolutely required. Notably feature flags triggering incidents is expected and desired vs the alternative of shipping the code and havi
429.
by baq 5mo ago
yes, worth every penny
430.
by baq 5mo ago
> it just reminds me on how feature flags can be misused as application configuration/customization. An antipattern i could observe at various organzations already. feature flags are perfect for configuration and customization, why
431.
by baq 5mo ago
National security, training data
432.
by baq 5mo ago
For the record I took it 100% seriously, having been wfh since covid I’ve taken more than one nap during ‘work hours’ myself… I’m saying ‘spin up an agent and go for a long walk’ being ok is close to over and it’s over in some companies alr
433.
by baq 5mo ago
What’s crazy is with the current gen of LLMs you could 100% rein in the abusive mods. SO could reinvent itself as a place for coding agents to get clankermotiomal support after they ‘hear’ ‘WHY THE F DID YOU DO THAT’. (Seriously a hook coul
434.
by baq 5mo ago
we'll have to agree to disagree I guess ;)
435.
by baq 5mo ago
Doesn’t matter as long as it isn’t ugly. Porsche made the cayenne and the panamera, too. The V12 buyer won’t even look at this, but the luxury EV buyer now has a new thing to consider.
436.
by baq 5mo ago
They notice, but aren’t inconvenienced (I guess? Don’t own one and haven’t seen one on a plane)
437.
by baq 5mo ago
All the ceremony must be replaced with process optimization, skill extraction, harness development and new model evals. Still better than dealing with people, but only just.
438.
by baq 5mo ago
Nap while you can. The baseline is slowly raising; AI fed with organization context will hunt you down and lay you off, as it has done at multiple companies this spring already.
439.
by baq 5mo ago
Tell that to my IT department please
440.
by baq 5mo ago
China can afford and has the political will and power to centrally plan parts of the economy it feels like planning. Cars are obvious examples and if dram is next, western manufacturers should brace for impact.
441.
by baq 5mo ago
The engineer is not the target user
442.
by baq 5mo ago
My coworkers don’t care, they’re doing the exact same thing.
443.
by baq 5mo ago
Sometimes I think folks having ‘experiences’ should play more poker. The point is, it’s a game of chance and yet good players beat bad players in the long run. Your job in the new era of software engineering is to design the process so LLMs
444.
by baq 5mo ago
‘If it hurts, stop’ vs ‘if it hurts, do more of it’. Organizations have a choice, some slow down to avoid, some speed up in hope to make issues… non-issues. If the go-fast orgs find workflows that actually truly speed things up without loss
445.
by baq 5mo ago
Price of PCs causing a collapse in demand, then mass bankruptcies of companies making PC components so supply chains get demolished and when prices come back down there’s no one left selling anything so you can’t build a pc at any price
446.
by baq 5mo ago
Wonder if LLMs in autoreasearch loops would be able to complete tasks geohot has in mind in say 100x average token budget. If the answer is yes, the argument doesn’t matter: you just run the loop and wait for llm analog of moore’s law to ge
447.
by baq 5mo ago
Would you ridicule Eddy Cue and Jeff Robin (heads of iCloud, kinda) for not running iCloud on macOS?
448.
by baq 5mo ago
That’s plenty good, it isn’t like anyone has the money to buy a new GPU
449.
by baq 5mo ago
If it crashes after it kills the PC we’ll be left with… nothing? Path matters as much as destination
450.
by baq 5mo ago
I hope they fail obviously, but I have to respect the exponential they’re betting on.

More ›