Skip to main content
Maarten Grootendorst

Maarten Grootendorst

Developer Relations Engineer at Google DeepMind and creator of BERTopic; author of "Hands-On Large Language Models" bestseller.

Aug 20, 2026

7:39
Hey, let me try these four different embeddings, right? So do you find that embeddings has a special place in that more people play around with embeddings and have some rudimentary understanding of embeddings?
7:51
Yeah, I think those are, to me, I have a sweet spot for embeddings because it's the main part of Bird Topic, right? But I think it's so fundamental to so many things that we do in this field.
8:04
Even like things like Rack, which some people think is outdated.
8:07
It actually isn't.
8:08
It's very much alive and still kicking.
8:11
Runs on embeddings and understanding how they work will also help you understand how LLMs work.
8:16
And it can be used in so many different ways.

21 MINS LATER

29:00
Gemma is not as good as Gemini.
9:38
four different embeddings, right? So do you find that embeddings has a special place in that more people play around with embeddings and have some rudimentary understanding of embeddings?
9:51
Yeah, I think those are, to me, I have a sweet spot for embeddings because it's the main part of Bird Topic, right? But I think it's so fundamental to so many things that we do in this field.
10:06
Even things like Rack, which some people think is outdated, it actually isn't.
10:10
It's very much alive and still kicking, runs on embeddings.
10:16
Understanding how they work will also help you understand how LLMs work.
10:20
And it can be used in so many different ways.
10:23
Sometimes we're looking for more bigger embedding models, contextualized information.

25 MINS LATER

35:31
Gemma is not as good as Gemini

We value your privacy

We use cookies to understand how you use our platform and to improve your experience. Click “Accept All” to consent, or “Decline non-essential” to opt out of non-essential cookies. Read our Privacy Policy.