Summary
Thomas Paul is an AI Compiler Engineer with a decade of software experience and over seven years focused on deep learning compiler frontends, INT8 quantization, and model deployment to edge devices. He has delivered production-ready reference runtimes and INT8 implementations across dozens of layers, enabling transformer and CV models to run with TFlite-compatible accuracy on custom hardware. Thomas has led and mentored small teams, built model zoos and tooling for compute/memory/accuracy analysis, and bridged sales and engineering through technical demos. Comfortable across the full ML lifecycle, he brings hands-on expertise in ONNX/PyTorch/TensorFlow conversions, calibration pipelines, and runtime packaging. Based in Dubai, he combines engineering rigor with a student mindset—continuously refining workflows through personal projects—and is as likely to debug quantization quirks as he is to be found reading fiction or following Formula One.
10 years of coding experience
15 years of employment as a software developer
High School/Secondary Diplomas and Certificates, High School/Secondary Diplomas and Certificates at Jawahar Navodaya Vidyalaya, Ernakulam
Nanodegree Self Driving Car, Nanodegree Self Driving Car at Udacity
Bachelor of Technology (B.Tech.) Computer Science and Engineering, Bachelor of Technology (B.Tech.) Computer Science and Engineering at Saintgits College of Engineering