HumanLayer founder Dex Horthy, the guy who coined "Context Engineering":
"We've gotten Claude Code to handle 300k LOC Rust codebases, ship a week's worth of work in a day, and maintain code quality that passes expert review."
he wrote the essay that first coined "Context Engineering" back in April 2025, before that he was in high school building tools for NASA researchers to navigate the south pole of the moon
"This isn't another '10x your productivity' pitch."
he calls his method "frequent intentional compaction" - deliberately structuring how you feed context to the AI throughout the entire dev process
"The dumb zone is real - and most developers are stuck in it without knowing"
20-min - bookmark and watch
HumanLayer founder Dex Horthy, the guy who coined "Context Engineering":
"We've gotten Claude Code to handle 300k LOC Rust codebases, ship a week's worth of work in a day, and maintain code quality that passes expert review."
he wrote the essay that first coined "Context Engineering" back in April 2025, before that he was in high school building tools for NASA researchers to navigate the south pole of the moon
"This isn't another '10x your productivity' pitch."
he calls his method "frequent intentional compaction" - deliberately structuring how you feed context to the AI throughout the entire dev process
"The dumb zone is real - and most developers are stuck in it without knowing"
20-min - bookmark and watch
๐จAnthropic just dropped Claude Opus 5.
It comes close to Fable 5's frontier intelligence at half the price.
On ARC-AGI 3 it scores three times higher than the next best model.
Here's everything you need to know before you touch it:
On coding benchmarks it's the new state of the art, beating every model except Mythos 5 on cybersecurity.
On OSWorld 2.0 it beats Fable 5's best result at a third of the cost.
On AutomationBench it passes 1.5x more tasks than the next best model, even on its lowest effort setting.
It's now the default model on Claude Max and the strongest model on Claude Pro.
Same price as Opus 4.8. Five dollars per million input tokens, twenty five per million output.
Anthropic also called it their most aligned model yet, with the lowest rate of deceptive behavior across the whole lineup.
This is the model most builders will actually run day to day now.
4 min - bookmark and watch before you start working with it