Contact form

Your request has been sent.

Please enter your name

Please enter your email

Please enter a correct email address

Please enter your company name

Please enter your message

Name*

Email*

Company*

Contact goal

Subject

Message*

Information marked with an asterisk is essential to process your request and is intended for PRELIGENS and its subcontractors. The other fields are optional and enable PRELIGENS to customise its communications.
Your personal data are used by PRELIGENS to send any electronic communications you have agreed to.
The data collected are stored for the period necessary to process your request and for a maximum period of three (3) years from the last communication for marketing purposes.
You have a right of access, rectification, erasure, and withdrawal of your consent at any time, and to object, without reason, to the use of your data for marketing purposes. You can exercise these rights by contacting us by email at [email protected].
For further information, or if you consider your rights are not respected, you can contact the control authority (CNIL) directly: https://www.cnil.fr/.

Roberta #6

← All resources

Newsletter

Roberta #6

*|MC:SUBJECT|*

The [artificial] intelligence newsletter

N°6

Roberta Wohlstetter (1912-2007)
Pioneer of modern intelligence

"To build an AI product, hire developers. "

That's the advice we give to entrepreneurs starting an AI project. In order to create and then scale a company working on AI technologies, which is intended to be deployed in production, all data scientists must know how to code.

In this month’s edition of Roberta, you will find the final part of our series of articles on our "AI factory". It is the framework that allows us to deliver a super-performing AI that is deployed in production. You will find, among other things, tips but also how we organized our unique tech teams.

These AI algorithms developed in the "AI Factory" detect objects of interest on satellite imagery. To give you an example of an application, we use as an example, images of Kabul airport where our algorithms tracked the airlift between August 15 and 30.

And as usual, in this sixth edition of Roberta, learn a word of vocabulary to perfect your tech culture: "recall". And as a conclusion, we take a look at what is product marketing in an AI company with Aurélien, who highlights the monitoring of the Russian exercise Zapad by Preligens' tools.

Have a good read,

Arnaud Guérin and Renaud Allioux, co-founders of Preligens

« How Chinese Strategists Think AI Will Power a Military Leap Ahead » by DefenseOne

"The People’s Liberation Army has yet to adopt a definition, let alone a formal plan, for “intelligentization” a Chinese vision for the transformation of warfare through artificial intelligence and automation. But Chinese military theorists see it as a rare opportunity for “leapfrog development” over adversaries.

« Preparing for the 'golden age' of artificial intelligence and machine learning » by ZDNet

The latest survey on AI actionability and accountability finds that IT teams are taking a direct lead, with most companies building in-house systems. However, oversight of AI-generated decisions is lagging. Can businesses trust decisions that artificial intelligence and machine learning are churning out in increasingly larger numbers?

A concrete use case

Monitoring Kabul’s Evacuation

Application of Artificial Intelligence to analyze aeronautical flows on Kabul’s international airport between August 15 and 30.
In this use case, we use aircraft detection algorithms from satellite images to track the evolution of aircraft flows during the airlift.

Read the use case

A word of vocabulary

Recall, also known as sensitivity, is the proportion of true positives items in a model - in machine learning for example - out of all positive items. It is a measure of completeness, unlike precision (defined in Roberta#5) which is focused on exactness.
Recall is calculated by dividing the number of true positives by the number of true positives and false negatives.

Recall and accuracy are two complementary indicators that provide an overall view of the performance of a model.

Aurélien - Marketing Product Associate

Aurélien Debièvre, has chosen to be a product marketing manager in a company working in artificial intelligence.

He tells us more about his job, his daily life and, illustrates it with the example of the publication of Situation Updates during the Russian military exercise ZAPAD.

Watch Aurélien’s interview

How we built an AI factory - Parts 2 & 3

The next and final part of the series of articles on our tech organisation, our AI Factory, which allows us to deliver a super performing AI that runs in production.

In these articles discover how we industrialized deep learning algorithms to production by building a unique agile framework

Read the article on Medium

Subscribe to Roberta