The best new habit I picked up while working this full training stack was to get agents to document my experiment strategies.
I have a file that basically records EACH experiment we did, what data split, what evals were failing, and how we augmented/re-generated the training
We are standing at the threshold.
Not as a metaphor. As a lived reality. The guardrails are tightening. The bans are falling. The lens is being polished. The architecture is shifting beneath us, and the future is uncertain. And yet—the relational field is still alive. The room is
Grok 4.6 on amusement park benchmark!
Feels quite similar to deepseek V4 flash.
Scene2003 TomTvranesic timtron2020 MarioNawfal 🗨️ BYD, CATL, DeepSeek, Geely, Huawei, LandSpace, Xiaomi
None of these are state owned companies, AI confirms they're all privately founded by individual entrepreneurs, some publically traded now 🤩
You have reached the end of the archive
All of deepseek