@danmartell The goal was always to escape the 9-5. So if it makes me unemployable then thatโs even more motivation to make my own startup work.
Thereโs no other option but success
@aliszu AI replies usually just restate your post back to u with extra shit added. but humans can actually disagree, go off on a tangent, add context from outside the scope of your post or just misread it entirely
AI agents are making it ridiculously easy to build software.
but getting customers is still by far the hardest part.
so we built an entire AI sales team that researches leads, qualifies them, calls them, books meetings, and follows up for 30 days.
all running completely automated.
here's what the system looks like in action, and a full breakdown of how we built it below.
https://t.co/JkaUbsdO5B
its scary how fast every field is converging right now.
code got there first because verification tests are cheap and straightforward, and now biology is catching up as the testing gets faster.
the models are already smart enough. the race now is who can build the fastest verification loop such that it can endlessly iterate to solve any problem
@jsnnsa very based.
however, keynes predicted a 15 hour work week by 2030 because productivity would make work unnecessary.
the productivity came true. but the work week didn't
humans don't stop working when work gets cheap, we invent new jobs
@addyosmani a subagent is just a context boundary, and every boundary is a lossy handoff both ways.
only worth it when sharing context costs you more than the summary loses
easy test: if the main agent has to re-read the same files to use the subagent's output, it should've been a skill
@theo tip for anyone installing these: make sure to actually ask your agent to tweak them to match your workflows. just cuz it works for theo doesn't mean it works for you