Guy Nicholson
Limited Partner at R42 Group
San Francisco Bay Area United States
Summary
🤩
Rockstar🎓
Top SchoolGuy Nicholson is a technology executive and Limited Partner based in the San Francisco Bay Area with a proven track record of bringing market-leading consumer devices to market for Google, Apple and Nokia. He combines deep expertise in embedded applications, audio systems and machine learning—work that includes hands-on contributions to Google Research models such as keyword spotting with quantization-aware training. Guy has led cross-continental product and engineering teams across Asia, Europe and North America, marrying product management discipline with acoustics expertise from a PhD in the field. His background ranges from device-level technology leadership to ML model optimization, giving him rare end-to-end fluency in both hardware and software productization. Known for translating research into production-ready systems, he thrives on solving acoustics and audio challenges at scale for consumer devices.
7 years of coding experience
18 years of employment as a software developer
PhD Acoustics, PhD Acoustics at The University of SalfordHigh School A and O levels, High School A and O levels at Sherborne School, UK
French
Github Skills (8)
mask-rcnn10
keras10
faster-rcnn10
machine-learning10
tensorflow10
python9
ai9
mlops5
Programming languages (7)
Github contributions (5)
google-research/google-research
Jun 2021 - Oct 2022
Google Research
Role in this project:
ML Engineer
ML EngineerContributions:14 commits in 1 year 4 months
Contributions summary:Guy primarily contributes to the model development and optimization within the Google Research repository, specifically focusing on keyword spotting. Their work includes updating models based on research papers and adding support for quantization aware training, notably through the implementation of n-bit quantization schemes. They also address internal changes related to model architecture and fix quantization-related issues.
googlemachine-learningai
guynich/distil-whisper
Feb 2024 - Mar 2024
Distilled variant of Whisper for speech recognition. 6x faster, 50% smaller, within 1% word error rate.
Contributions:56 PRs, 51 pushes, 64 branches in 27 days
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.