Sander Schulhoff
speaker
947 appearances
1 recordings
1 series
first heard Jun 2025
last heard Jun 2025
Sander Schulhoff’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsNo recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.
Appearances
I don't know.
It's just one of those kind of random LLM things.
But I had to add in that uh thought-inducing phrase like, you know, make sure to write out all your reasoning in order to make sure that happens.
Because I I wanted to make sure to maximize my performance over my whole test set.
So what we see is that you know new model comes out, people are like, ah, you know, it's so good.
You don't you don't even need to prompt engineer it.
You don't need to do this.
But if you look at scale, if you're running thousands, millions of inputs through your prompt, uh oftentimes in order to make your prompt more robust, you'll still need to use those classical prompting techniques.
I'd say Okay.
No need, but if you're using GPT four, GPT four oh, then it's still worth
Yeah.
I I guess just quickly maybe a maybe a reality check is like the way that I do kind of regular conversational prompt engineering is
I'll just be like, you know, if I need to write an email I'll just be like
Ret email, like not even spelled properly, uh, about you know, about whatever.
I usually won't go to all the effort of showing it my previous emails.
Uh and there's a lot of situations where I'll you know, I'll paste in some writing and just be like, make better, improve.
Uh so that like super, super short uh
lack of details, lack of any prompting techniques, that is the reality of a large part, the vast majority of the conversational prompt engineering that I do.
There are cases that I will bring in those other techniques, but
the most important places to use those techniques is the product focused prompt engineering.
Showing 421–440 of 947 · page 22 of 48
← Previous
Next →