the gap shows up later because seniors are just juniors who got reps
cut the reps and the pipeline dries up
will companies hire juniors back in 2 years or just pay seniors more?
fewer subagents makes sense, every handoff loses context
one agent with a clear task usually beats five that guess
what's your cutoff, 2 agents or 3?
the tap a balloon and see the energy flow part is what makes this work
static diagrams never let you poke the system
would this style work for chemistry too?
claude fixed bad speaker sound from a mic and a prompt
this is the quiet shift: software fixing hardware problems
what's the first device you'd hand to an ai to fix?
agents tapping through the ios simulator is the missing half of mobile dev. the hard part will be flaky animations and permission popups, not the happy path
json scenes make sense for agents. the open question is whether it can playtest its own game and catch what feels bad, not just what fails a schema check
types as feedback is underrated. a compile error is a free, instant reviewer the agent can't argue with
claude writes about 90 percent of claude code, and now the review process is the real product
when the model ships most of the code, the human job turns into deciding what's safe to merge
would you trust ai reviewing ai code, or does a human always sign off?
if terra really undercuts sol on price and wins on quality, the pricing pressure on everyone else gets real. rumor only so far though
anything that is just a thin ui over a database is exposed. the safe ones own data, distribution or compliance that a model cant copy
open weights at that price is the real story. curious if the medical scores hold up outside the benchmark
auditing your own session logs is the underrated fix for usage limits. my guess is the biggest burn is re-reading huge files in long sessions, curious if that shows up as habit one
3.6x token volume in a week is a real number. the open question is whether those users stay once the next openai model ships, or if this is just launch week churn
driving unreal fully from the cli with claude code is the interesting part. how are you checking the result without opening the editor, screenshots from a headless run?
tvs that record audio and screenshots for ads is the best argument for local ai
if your tv can be a spy, your ai assistant sure can
would you pay more for a tv with zero cloud, or is this already normal?
