Site Reliability Engineer at Veesion

Site Reliability Engineer at Veesion

Job Description

We are looking for a Site Reliability Engineer to help us deploy our algorithms in retail stores. Our focus is on efficient, easy-to-scale, easy-to-maintain on-premise deployments.

-Automate the server provisioning process.
-Research the most efficient solution for integration (Jetson, FPGA, GPU server).
-Plug the solution to local Digital Video Recorder (DVR) or directly to IP cameras.
-Develop tooling to monitor systems and diagnose performance bottlenecks.
-Eliminate inefficiencies and simplify our build process.

Requirements

-Linux (incl. Bash).
-Databases (preferably PostgreSQL or similar).
-Experience with one or more of these languages: Python, Go, Java, C/C++, C#.
-Experience with OpenCV is a big plus.
-Experience in video network setup.
-Configuring and maintaining secure network architecture.
-Knowledge of resource management and job scheduling tools (SLURM, SGE, etc.)

Bibliographie

[1] Bolei Zhou, Alex Andonian, Antonio Torralba : Temporal Relational Reasoning in Videos arXiv:1711.08496

[2] Jiawei He, Mostafa S. Ibrahim, Zhiwei Deng, Greg Mori : Generic Tubelet Proposals for Action Localization arXiv:1705.10861

[3] Sijie Yan, Yuanjun Xiong, Dahua Lin : Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition arXiv:1801.07455

[4] Max Jaderberg, Karen Simonyan, Andrew Zisserman, Koray Kavukcuoglu : Spatial Transformer Networks arXiv:1506.02025

[5] Zhenzhong Lan, Yi Zhu, Alexander G. Hauptmann : Deep Local Video Feature for Action Recognition arXiv:1701.07368

[6] Eddy Ilg, Nikolaus Mayer, Tonmoy Saikia, Margret Keuper, Alexey Dosovitskiy, Thomas Brox : FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks arXiv:1612.01925

[7] Thomas Schlegl, Philipp Seeböck, Sebastian M. Waldstein, Ursula Schmidt-Erfurth, Georg Langs : Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery arXiv:1703.05921

[8] https://github.com/TwentyBN/GulpIO

Office

Paris – Full time

Compensation

€40k – €80k . 0.1% – 0.5%

About Veesion

Founded by Polytechnique and HEC alumni, Veesion is incubated at Agoranov (96 bis Boulevard Raspail, 75006 Paris).

We are developing a gesture recognition technology in video content. The potential applications are diverse and our first focus is shoplifting detection in the retail industry.

We built a large video database thanks to partnerships with major retailers and we are in constant discussion with top-notch Computer Vision labs such as the Thoth and Willow teams in order to drive our research.

We are a young and agile startup: joining us now is the opportunity to grow with the company.

Deep Learning Engineer at Veesion

Deep Learning Engineer at Veesion

Job Description

The mission mainly consists in improving the accuracy of the algorithm we are developing on a proprietary video dataset. Tasks will include:

-Tracking down and implementing state-of-the-art video recognition deep learning architectures [1-2-3-4]
-Implementing a light optical flow based action recognition model [6]
-Designing intermediary descriptors [4-5]
-Designing an unsupervised approach to the problem [7]
-Mastering general data preprocessing techniques [8] and temporal data augmentation
-Implementing an efficient pruning for all neural networks used
-Improving the multithreading and multiprocessing training and inference pipelines
-Collaborating with the Integration team and the Product Development team
-Assisting the CTO in tech related appointments and collaboration with labs
-Explore new tasks related to the algorithm arising from tech watch

Requirements

– Proven expertise in Deep Learning, especially in computer vision (video related tasks are a plus)
– Interest in research and ability to extract useful information from scientific papers
– Solid grasp of at least one of the following deep learning frameworks : Pytorch, Tensorflow, CNTK, Caffe
– Understanding of convolution and famous related architectures (resnext, I3D, RPN, two-stream networks)
– Analytical mind, ability to take a step back and see the big picture
– Problem-solving aptitude
– Master / PhD in Deep Learning or relevant fields.

Bibliographie

[1] Bolei Zhou, Alex Andonian, Antonio Torralba : Temporal Relational Reasoning in Videos arXiv:1711.08496

[2] Jiawei He, Mostafa S. Ibrahim, Zhiwei Deng, Greg Mori : Generic Tubelet Proposals for Action Localization arXiv:1705.10861

[3] Sijie Yan, Yuanjun Xiong, Dahua Lin : Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition arXiv:1801.07455

[4] Max Jaderberg, Karen Simonyan, Andrew Zisserman, Koray Kavukcuoglu : Spatial Transformer Networks arXiv:1506.02025

[5] Zhenzhong Lan, Yi Zhu, Alexander G. Hauptmann : Deep Local Video Feature for Action Recognition arXiv:1701.07368

[6] Eddy Ilg, Nikolaus Mayer, Tonmoy Saikia, Margret Keuper, Alexey Dosovitskiy, Thomas Brox : FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks arXiv:1612.01925

[7] Thomas Schlegl, Philipp Seeböck, Sebastian M. Waldstein, Ursula Schmidt-Erfurth, Georg Langs : Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery arXiv:1703.05921

[8] https://github.com/TwentyBN/GulpIO

Office

Paris – Full time

Compensation

€40k – €80k . 0.1% – 0.5%

About Veesion

Founded by Polytechnique and HEC alumni, Veesion is incubated at Agoranov (96 bis Boulevard Raspail, 75006 Paris).

We are developing a gesture recognition technology in video content. The potential applications are diverse and our first focus is shoplifting detection in the retail industry.

We built a large video database thanks to partnerships with major retailers and we are in constant discussion with top-notch Computer Vision labs such as the Thoth and Willow teams in order to drive our research.

We are a young and agile startup: joining us now is the opportunity to grow with the company.

Le Parisien – “Sceaux: Veesion technology will help retailers to detect shoplifters”

“Shoplifting is a scourge for the retail industry and only 5% of thefts are  detected by the current surveillance solutions,” says Thibault David, 25, one of the co-founders of Veesion. A team of former students graduated from HEC Paris and Polytechnique, has developed a technology which detects shoplifting gestures on security cameras, thanks to an artificial intelligence algorithm.

Check this new article about Veesion in Le Parisien.

http://m.leparisien.fr/hauts-de-seine-92/sceaux-leur-logiciel-permettra-de-reperer-les-voleurs-dans-les-magasins-05-10-2018-7911960.php

Veesion at “L’Odyssée du CDSE LAB”

 

On September 27th, we were proud to have welcomed Gerard Collomb, the Minister of Internal Affairs, for a live demo of our anti-shoplifting solution.

 

We also had the pleasure to discuss the potential applications of our gesture recognition technology in the security / defense sector.

 

A huge thank you to CDSE, one of the top AI & Security exhibition.

Veesion is part of the 5th batch at Lafayette Plug & Play accelerator!

Lafayette Plug and Play is an international business focused catalyst for startups in the retail, e-commerce and fashion tech industries.

Veesion will benefit from a 3-month acceleration program focused on business development during which startups boost sales opportunities with a wide global network of corporate partners (top retailers, brands & e-merchants).

https://business.lesechos.fr/entrepreneurs/actu/0302199335430-lafayette-plug-and-play-lance-sa-cinquieme-promotion-323017.php

A big thank you to Les Echos for the recent article.