Chaitanya Asawa
① 他说过的话
8 条
对我们来说,我们真正把 evals 当作操作系统,当作公司的命脉。
指向原始笔记的链接
For us, we really treat evals as the operating system, the life’s blood of the company.
—— Chaitanya Asawa · [12:47]
但在医疗领域,如果我们答错了什么,是会有真实后果的,我们会完全失去信任。
指向原始笔记的链接
But in healthcare, if we answer something incorrectly, there’s actually consequences, and we entirely lose our trust.
—— Chaitanya Asawa · [11:57]
一切都围绕对话,那场神圣的医生与患者的对话,而我们只是围绕它构建了这一整套行政机器。
指向原始笔记的链接
It’s all about the conversation, that sacred doctor and patient conversation, and we’ve just built all this administrative machinery around that.
—— Chaitanya Asawa · [08:57]
如果我有一个非常非常好的生成器验证器,那它本身就是我的生成器了。
指向原始笔记的链接
If I had a really, really good generator, verifier, then that would just be my generator itself.
—— Chaitanya Asawa · [16:08]
我认为一个真正出色的评估系统有一个特性:它反映了你在产品中想要的行为。
指向原始笔记的链接
You know, I think a really great evaluation system has a property that it reflects the behaviors that you want in your product.
—— Chaitanya Asawa · [14:08]
我们不是每个问题都需要前沿级别的智能。
指向原始笔记的链接
We don’t need frontier-level intelligence for every problem.
—— Chaitanya Asawa · [18:49]
我们的关键洞察是,如果我们拥有制胜权——也就是拥有他们可能没有的合适数据,以及聚焦于他们可能没有聚焦的问题——我们实际上有可能跑赢前沿模型的变化速度。
指向原始笔记的链接
Our key insight is we can actually potentially beat the rate of change on the frontier model if we have the right to win by having the right data that they may not have and the focus on a problem that they may not be focusing on.
—— Chaitanya Asawa · [19:33]
再说一次,利害真的非常高,你不能只是像个原型那样直接发出去,然后说,yeah,我在几个案例上测试过,它没问题,这样是行不通的。
指向原始笔记的链接
Again, the stakes are really high, and you cannot get away with just being like a prototype that you just ship out there and be like, yeah, I mean, I tested on a few cases and it works.
—— Chaitanya Asawa · [13:39]
② 出现在这些集
1 集
③ 他谈到的
点进去有真内容 —— 本页主要出口
Abridge · Glean · 智能体 · 临床决策支持 · 评估 · LLM 评委 · 评分细则 · 后训练 · 前沿模型 · 临床记录
