this is a really cool idea, BUT... 

for every "superbuilder" a team has focused on build agents and streamlining team ops with LLMs... 

you'll also need a "context designer" or "librarian" making sure that the underlying data and information architecture is sound enough to get the most out of these tools

tools are everywhere; software is easy to build 

but using LLMs at scale makes it infinitely more difficult to maintain coherence and consistency across internal context and external narratives 

without someone focused on that full-time, your team's AI empire is a house of cards

🎉 Just announced -- Superbuilder role 🎉

We are launching a new role on @base engineering today -- it's called Superbuilder.

Become Superbuilder #1 and apply. Read more in 🧵 https://t.co/GDn5LQU0da

Chintan Turakhia🛡️

this is a really cool idea, BUT... 

for every "superbuilder" a team has focused on build agents and streamlining team ops with LLMs... 

you'll also need a "context designer" or "librarian" making sure that the underlying data and information architecture is sound enough to get the most out of these tools

tools are everywhere; software is easy to build 

but using LLMs at scale makes it infinitely more difficult to maintain coherence and consistency across internal context and external narratives 

without someone focused on that full-time, your team's AI empire is a house of cards

https://x.com/chintanturakhia/status/1915866734437056911

seeker /native /believein /bedrock 🇵🇸

Solid point. LLMs are powerful, but without strong data architecture and someone constantly curating context, all that potential can easily get lost in translation. 

Having a "context designer" seems like the missing link to avoid chaos as AI scales.