I'm an AI Architect and Researcher with a relentless pursuit of innovation. In Google Research, I've pioneered the development of Generative Models for synthetic human creation for the Pixel FaceID. I've designed, developed and deployed a model that could render different human faces of different ethnicities and under different field conditions to improve our existing facial recognition models. As a result, this model improved Facial Recognition performance for underrepresented ethnicities by an order of magnitude while saving over 80M USD in data collection costs.
Leveraging my experience at Google Research, I spearheaded the development of groundbreaking virtual try-on technology at Korgi.ai, developing Hypernetwork based Generative models that could render photorealistic clothing from a brand's catalog into customer's bodies, or even put them in AI created synthetic models. This technology was leveraged to reduce marketing costs in promoting fashion lines while increasing conversions. In this company, I served as the Founder and Principal research scientist, managing a cross functional team of 10. I also worked as an AI team lead for Smile Identity, a Series B biometrics startup operating in Africa. In there, I've managed a team of 5 consisting of research scientists and developers. I've designed custom models to prevent fraudulent biometrics attacks, developed custom low footprint facial recognition algorithms that can function in low tech settings and architected the infrastructure to enable this service to serve millions of requests. I'm now working on Synthhaven, a company that uses small, independent LLM agents to automate the creation of marketing campaigns, advertising campaigns. We are building a suite that lets agents interact with a business' entire marketing pipeline and iterate on different campaigns.

Ralph Blanes

AI Architect | Founder @ Synthhaven | Ex Google Brain | LLM and Diffusion Models Expert

About

I'm an AI Architect and Researcher with a relentless pursuit of innovation. In Google Research, I've pioneered the development of Generative Models for synthetic human creation for the Pixel FaceID. I've designed, developed and deployed a model that could render different human faces of different ethnicities and under different field conditions to improve our existing facial recognition models. As a result, this model improved Facial Recognition performance for underrepresented ethnicities by an order of magnitude while saving over 80M USD in data collection costs. <br> Leveraging my experience at Google Research, I spearheaded the development of groundbreaking virtual try-on technology at Korgi.ai, developing Hypernetwork based Generative models that could render photorealistic clothing from a brand's catalog into customer's bodies, or even put them in AI created synthetic models. This technology was leveraged to reduce marketing costs in promoting fashion lines while increasing conversions. In this company, I served as the Founder and Principal research scientist, managing a cross functional team of 10. I also worked as an AI team lead for Smile Identity, a Series B biometrics startup operating in Africa. In there, I've managed a team of 5 consisting of research scientists and developers. I've designed custom models to prevent fraudulent biometrics attacks, developed custom low footprint facial recognition algorithms that can function in low tech settings and architected the infrastructure to enable this service to serve millions of requests. I'm now working on Synthhaven, a company that uses small, independent LLM agents to automate the creation of marketing campaigns, advertising campaigns. We are building a suite that lets agents interact with a business' entire marketing pipeline and iterate on different campaigns.

Education

Georgia Institute of Technology

Bachelor of Science / May, 2018

Atlanta, Georgia, United States of America

Experience

Google

Research Engineer / June, 2018 — December, 2021

Designed and developed GAN based methods to create synthetic human data for the Pixel FaceID, improving performance on underrepresented ethnicities, saving around 80M USD in data collection costs. Developed and ported on device CNN models for biometric fraud detection to protect users from having their phone unlocked by malicious actors. Models increased protection rate against cheap, easy to do attacks by an order of magnitude. Researched and implemented XAI research approaches for Google Cloud partners and Google Search as part of Google Brain, implementing advanced interpretable AI methods to facilitate decision making and confidence around deploying large AI models.

Join Ralph on NotedSource!

Join Now

At NotedSource, we believe that professors, post-docs, scientists and other researchers have deep, untapped knowledge and expertise that can be leveraged to drive innovation within companies. NotedSource is committed to bridging the gap between academia and industry by providing a platform for collaboration with industry and networking with other researchers.

For industry, NotedSource identifies the right academic experts in 24 hours to help organizations build and grow. With a platform of thousands of knowledgeable PhDs, scientists, and industry experts, NotedSource makes connecting and collaborating easy.

For academic researchers such as professors, post-docs, and Ph.D.s, NotedSource provides tools to discover and connect to your colleagues with messaging and news feeds, in addition to the opportunity to be paid for your collaboration with vetted partners.