George Gekov

I'm a

About

George Gekov is a Senior ML Applications engineer in Arm, the world's leading Semiconductor IP company.

Responsible for helping Arm partners run ML workloads on low-power CPUs and ML co-processors, George has a breadth of experience in compiling, optimizing & deploying neural networks on embedded microprocessors.

Key information

  • Technology area: Embedded Machine Learning
  • Specialty: Accelerating ML inference, Neural Networks optimisation
  • Data: Video, Voice, Vibration
  • Engineering Degree: Sorbonne University
  • Business Degree: Cambridge Judge Business School
  • Languages: C, C++, Python
  • Languages: English, French, Bulgarian

Skills

George has been fortunate to be working at the cutting edge of Machine Learning & low-power embedded technology for the past 5+ years.

ML inference acceleration100%
Deep Learning architectures80%
ML benchmarking75%
Neural Network optimisations 75%
Algorithm design 80%
Embedded programming 85%

Resume

Sumary

George Gekov

  • Making Machine Learning products a reality.
  • Customer-facing ML engineer.
  • Excellent communication skills.
  • Speaks fluently English, French, Bulgarian.
  • Presenter at International Tradeshows.

Education

Master in Entrepreneurship

2018 - 2020

Cambridge Judge Business School - University of Cambridge, UK

Courses include innovative business models, strategy, financial intelligence, assembling and managing high performing teams. The program is part-time, hence split in two years.

Master in Electronics & Computer Sciences

2014 - 2017

Polytech Paris Sorbonne, France

Program in a French École d'ingénieur focussing on analog & digital electronics and embedded systems.

Preperatory classes

2012 - 2014

Lycée Louis Le Grand & Lycée Pierre de Fermat, France

Advanced courses in Mathematics, Physics and General Engineering.

Professional Experience

Senior ML Applications Engineer

2022 - Present

Arm, Cambridge, UK

  • Help partners reach tape-out & mass market on ML-enhanced silicon devices.
  • Assist in choice of IP for partner SoCs to meet Power, Performance & Area requirements for ML workloads.
  • Optimising Neural Networks.
  • ML Performance Analysis - identifying bottlenecks in neural networks & making improvement recommendations.
  • Deliver ML on Arm training courses to partners from the Arm ecosystem.
  • Publish content(example 1,example 2).
  • Having fun & learning A LOT.

ML Software Engineer

2020 - 2021

Arm, Cambridge, UK

  • Development of embedded ML applications.
  • Example applications include keyword spotting, image classification, noise reduction.
  • Used TF Lite Micro inference engine & ArmNN.
  • Was part of a world class team & learnt a lot.

IoT Engineer

2017-2020

Arm, Cambridge, UK

  • Presenting Arm's IoT value proposition at Mobile World Congress.
  • Created a digital health demonstration - firmware updates over a cellular network on a connected medical device. Enables the sale of a medical cloud platform.
  • Led creation of a software translating demonstration between IP and non-IP protocols for a Linux gateway.
  • Extensive programming of Arm Cortex-M microcontrollers
  • Demonstrations of a cellular connectivity management portal showcasing RSP on eSIM and iSIM.

Systems Automation Intern

2017

Arm, Sophia Antipolis, France

  • Developement & intstrumentation of an internal IP verification flow.
  • Heavy usage of Python and data science technologies(Apache SPARK,HDFS).

Embedded Software Intern

2016

Arm, Cambridge, UJ

  • Profiling of automotive code on ARM Cortex-A and Cortex-M processors.
  • Used Performance Monitor Unit to measure number of CPU cycles, cache hit/cache misses ratios and branch predictions.

Contact

Send me a linkedIn request if you would like to get in touch. Please include reason you want to connect.