Ge: Neural network ad models run far more efficiently on modern GPUs than decision trees
Gio (Giovanni) · Inside AppLovin’s $100B Ad Engine · Aug 14, 2026 · at 16:42
AppLovin CTO Gio Ge explains why deep neural networks proved cheaper to inference on modern GPU hardware than legacy decision trees.
0:00 / 0:30exact quote · 30.0s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“The neural network, the benefit of neural network is, it's made up of a lot of standard GMM, general matrix multiplication, which, you know, is a famous, the GPU. GPU is highly optimized for this kind of operation. But the selection tree, you know, the old generation selected trees has Like has the structure that is very hard to optimize on a GPU machine. So the new generation model is very, very powerful and efficient with a modern GPU architecture.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Gio (Giovanni)
Gio: Wasteful ideas and bad taste offset AI's productivity gains
“I think a big, big mistake that people are making is that because of the AI, building things become easier. People are making a lot of bad decisions, bad tastes, wasteful ideas that actually offset the empowerment of AI.”
AppLovin strictly holds engineers accountable for any AI-generated code decisions
“You can use AI however you want.
But in the end of the day, you as a human is held accountable for every decision that your AI made for you.
So it is absolutely not acceptable when I come to an engineer and say, hey, why you did, why you implemented this syste…”
Gio: AppLovin is developing fully automated AI ad creative generation
“We have to create a tool that can fully automate everything. The other choice is you just create a tool that allows advertisers to interact with, to allow them to collaborate with the AI tool to create their own content. So right now we see that the second typ…”