increasing levels of autonomy: /skill: preset prompts /plan: human-refined inputs /goal: AI-evaluated outputs
@swyx
-
Human-AI Bandwidth Bottleneck in Modern ML Accelerators
By
–
In modern ML accelerators, FLOPS have absolutely exploded. Often though, the bottleneck is not FLOPS but memory bandwidth. Similarly, model intelligence has exploded, causing the bottleneck to be humanAI bandwidth. At Thinky, we think that it’s important to solve this. 1/4 x.com/thinkymachines…
-
AI Performance Leap: ‘Framemogged’ Realtime Capabilities Redefined
By
–
I believe the kids call this "
@thinkymachines just brutally framemogged gdm and oai". basically everyone's definition of "realtime" just got a massive frciking upgrade -
OpenAI’s New Venture Faces Server-Side Rendering Issues
By
–
openai's new $10b forward deployed company is so locked in they can't even bother to server side render properly (jk congrats, nice win for UK AI )
-
Distinguishing Orchestrator and Harness in AI Task Completion
By
–
unclear the diff between orchestrator and harness since they both have TASK_DONE
-
Essential AI Learning Resource for Engineers Discussed
By
–
this is a big deal, on the order of Kelsey Hightower’s “Kubernetes The Hard Way” and probably all ai engineers should go thru this once mostly i advocate “just in time learning”, but this is one scenario you want “just in case”
-
Team’s Success with Devin 3.0 and Devin Desktop Highlighted
By
–
i'm just an advisor, cant take credit. team is doing great with devin 3.0 and devin desktop
-
OpenAI vs Anthropic Valuation and ARR Comparison 2024
By
–
OAI 850B valuation, ~30B ARR now Ant ~900B valuation, ~44B* ARR now *revenue recognized differently, per Denise Dresser its probably closer to 8-10B lower if using OAI methodology chart reconstructed from wsj by me
-
OpenAI’s Testing Failures Raise Serious Safety Concerns
By
–
goblingate + glazegate casts serious doubt on whether or not oai does literally ANYthing beyond employee vibechecks before shipping this stuff (and even with minor fails theyre fine prompt tuning it lol)
-
Vibe-Kanban Shuts Down Live Onstage with 30K MAU, Goes Open Source
By
–
Much respect to @tokengobbler who shutdown Vibe-kanban live onstage at AIE Europe – still with 30,000 MAU, and still living on as an open source project. "Everyone who is making money is doing 2 things: selling to enterprise, and reselling tokens. We were doing neither."