Hicham Badri

speaker

223 appearances 2 recordings 1 series first heard Jun 2026 last heard 14 Jul

Hicham Badri’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

Recordings per month over the last 12 months — 2 in all, peaking in Jul 2026 with 1.

Appearances

newest first · ▶ plays the moment

Working Smarter · Building AI that can search inside videos (and photos and audio too) · 14 Jul 2026 podcast

The scene has ended here.

Now this is another scene.

The approach we are following right now, which is like the easier way of doing this, is separating modalities.

It's just scales much better.

Um and it's easier to serve and for inference cost.

Because you're basically decoupling things.

It's not the best way to do it.

But it's a valid way to do it at scale because

when you're separating this these components, you can also iterate very quickly.

You can, let's say, change like there's a new model, let's say, and the sense image is much better, you can just change that part.

There's a new transcription model which works very good, works with more languages.

We can just swap that.

And then we have we don't have to deal with all the headache of changing everything.

So the current approach we are following is very good in terms of scaling and iterating, but it's not the s best way

to have like full understanding because right now, for example, we see models that

They are called Omni Models.

She's you have one single model that that does everything actually.

It understands everything.

So these omni models, they are really, really interesting because they can like you can pretty much ask them every anything and you can give them like the audio and the images.

And they would make sense of everything.

Showing 61–80 of 223 · page 4 of 12