Maxime Portaz

Version Française

Machine Learning Engineer

Paris, France




Deep Learning Frameworks:

  • Pytorch
  • Tensorflow
  • Jax

Programming Languages:

  • Python
  • C++
  • C#
  • Swift


Field AI Engineer - Graphcore
Sept. 2021 - Current

Graphcore - Paris/Bristol
Help and support client in IPUs usage and deployment.

Languages : Python
Fields : Machine learning, Deep Learning, Computer vision, Natural Language Processing, etc.
Tools : Pytorch, Tensorflow, Jax, Poplar

Research Scientist - Trax Retail
May 2020 - Sept 2021

Trax Retail - Paris

  • Machine learning for frictionless stores
  • Neural networks on smartphone (IPhone and IPad) for object detection
  • Detection and people tracking in multi-camera stores
Models : CNN, RNN, LSTM, Fast-RCNN, et Yolo.
Languages : Python, NodeJS, Swift
Fields : Machine learning, Deep Learning, weight sensors, people tracking, and action recognition
Tools : Pytorch, OpenCV, Tensorflow, AWS

Research Scientist and Product Owner - Qwant Research
Nov. 2018 - Mai 2020
Qwant Research team, working on Image Search.
  • Deep Learning for image representation
  • Text and Image Classification
  • Graphcore IPUs test (link)
  • Scientifique publication of a new approach for Image Search Engines(link)
  • Product Owner of QISS (Qwant Image Similarity Search)
Languages : Python, C++, Javascript
Fields : Image indexing, Deep Learning, Natural Language Processing, text/document representation and query classification.
Tools : Pytorch, Vespa, Kubernetes, NVidia DGX, Microsoft Azure, Graphcore IPUs

PhD candidate - Laboratoire d'informatique de Grenoble
Sept 2015 - Oct. 2018

  • Scene and gesture recognition from egocentric vision
  • Learn to rank for images and instances
  • Transfer learning
Languages : Python, C++
Fields : Image indexing, Deep Learning, image representation, action recognition
Tools : Pytorch, OpenCV, Caffe

Research Engineer - Inria
Nov. 2014 - Août 2015

Pervasive Interaction - Inria Grenoble.

  • Narrative Design with Figurines : video
  • Multimodal data
  • 3d engine
  • Motion Capture with Kinect
Languages : C++
Fields : 3D models, scene generation, data fusion
Tools : OpenGL, Kinect SDK.


Université Grenoble Alpes
2015 - 2018

Doctorat - Recherche d'information multimédia

Ensimag - UGA
2013 - 2015

MOSIG Master Of Science In Grenoble - Graphismes, Vision et Robotique

Université Joseph Fourier
2010 - 2013

Licence - Mathématiques et Informatique