[ 66616680.000 ] posts.x: Full talk on Antigravity's sub-agent swarms, sidecars, and generative UI, including the OS kernel built by 93 sub-agents:  |   [ 66616680.000 ] posts.x: Give Messi the ball and get out of the way. That's the coaching logic @kevinhou22, engineering lead for Antigravity at Google DeepMind, applies to AI…  |   [ 66607680.000 ] posts.x: Full talk on treating a fleet of coding agents like units in an RTS, and what that changes about who burns out:  |   [ 66607680.000 ] posts.x: Scaling up coding agents doesn't remove the bottleneck, it just moves it onto you: more agents means more steering, more reviewing, more approving…  |   [ 66598380.000 ] posts.x: Full talk on Conductor's six principles, including the CIA database and the cloud sandboxes:  |   [ 66598380.000 ] posts.x: Most engineers hear about the best AI coding workflows through their social graph, three to six months after everyone else already knows…  |   [ 66541620.000 ] posts.x: @yoheinakajima  |   [ 66512340.000 ] posts.x: Full talk on the RMSNorm paper, the CUDA streams bug, and getting it running in production:  |   [ 66512340.000 ] posts.x: A race condition between two CUDA streams once made a language model repeat itself mid-sentence and lag a step behind its own output. @f_makraduli…  |   [ 66502500.000 ] posts.x: Full talk on why robotics still has no internet-scale dataset, and Skild AI's omnibodied answer to that gap:  |   [ 66502500.000 ] posts.x: A robot that could look at a pattern of stacked blocks and rearrange loose ones to match it, by sight, dates to the 1960s. A teleoperated…  |   [ 66491520.000 ] posts.x: Full talk on how Dyna turned reward models and active learning into a 99.4% success rate on napkin folding:  |   [ 66491520.000 ] posts.x: A robot that succeeds 80 to 90% of the time on a task will string together ten flawless repeats less than 0.1% of the time. @JasonMa2020, CTO and…  |   [ 66436200.000 ] posts.x: Full talk on why Moonlake builds its simulations from causal, code-based object models instead of generated video:  |   [ 66436200.000 ] posts.x: A photorealistic walkthrough of a room and a simulation you can actually act inside of are not the same thing, and mixing them up is the mistake…  |  

[ 66616680.000 ] posts.x: Full talk on Antigravity's sub-agent swarms, sidecars, and generative UI, including the OS kernel built by 93 sub-agents:  |   [ 66616680.000 ] posts.x: Give Messi the ball and get out of the way. That's the coaching logic @kevinhou22, engineering lead for Antigravity at Google DeepMind, applies to AI…  |   [ 66607680.000 ] posts.x: Full talk on treating a fleet of coding agents like units in an RTS, and what that changes about who burns out:  |   [ 66607680.000 ] posts.x: Scaling up coding agents doesn't remove the bottleneck, it just moves it onto you: more agents means more steering, more reviewing, more approving…  |   [ 66598380.000 ] posts.x: Full talk on Conductor's six principles, including the CIA database and the cloud sandboxes:  |   [ 66598380.000 ] posts.x: Most engineers hear about the best AI coding workflows through their social graph, three to six months after everyone else already knows…  |   [ 66541620.000 ] posts.x: @yoheinakajima  |   [ 66512340.000 ] posts.x: Full talk on the RMSNorm paper, the CUDA streams bug, and getting it running in production:  |   [ 66512340.000 ] posts.x: A race condition between two CUDA streams once made a language model repeat itself mid-sentence and lag a step behind its own output. @f_makraduli…  |   [ 66502500.000 ] posts.x: Full talk on why robotics still has no internet-scale dataset, and Skild AI's omnibodied answer to that gap:  |   [ 66502500.000 ] posts.x: A robot that could look at a pattern of stacked blocks and rearrange loose ones to match it, by sight, dates to the 1960s. A teleoperated…  |   [ 66491520.000 ] posts.x: Full talk on how Dyna turned reward models and active learning into a 99.4% success rate on napkin folding:  |   [ 66491520.000 ] posts.x: A robot that succeeds 80 to 90% of the time on a task will string together ten flawless repeats less than 0.1% of the time. @JasonMa2020, CTO and…  |   [ 66436200.000 ] posts.x: Full talk on why Moonlake builds its simulations from causal, code-based object models instead of generated video:  |   [ 66436200.000 ] posts.x: A photorealistic walkthrough of a room and a simulation you can actually act inside of are not the same thing, and mixing them up is the mistake…  |