Straight to the Decision Point

Writing is the earliest storage system

No information, no intelligence

No decision is a bad decision

No goal focus is ressource gluttony

These are true for both human beings and artificial intelligence algorithms.

At, we build solutions helping decision makers straight to the decision point.

#ai #decisionmaker #productivity

Speed and simplicity

According to Reid Hoffman, co-founder and chairman of LinkedIn, these two basic rules help enable decision making in complex environments:

  1. Speed
    Making quick decisions is a key success factor, provided that you have enough information to back them up with the right arguments. Knowing how to prioritize and rank arguments is an essential step in decision making, but the criteria for choosing between them are sometimes so numerous that it is difficult to arbitrate. This is why it is necessary to know how to simplify things.
  2. Simplicity
    Knowing how to simplify the data of a problem is the other key factor that allows leaders to make a difference. It is the simplification of a problem that allows speed. In modern organizations, the flow of information is incessant and saturates our analytical capabilities. Immediate access to key elements allows us to make quicker and more informed decisions.

Our approach is based on these two principles. helps Directors and Managers to make the right decisions faster.

#ai #productivity #ceo #csuite #decisionmaker #decisionmaking

Melting down gigantic GPU and CPU hardware is not philosophy for building products

#hardware #ai

Hopes in quantum computing announce game-changing improvements in artificial intelligence (AI) embedded in consumer-life products. AI research is moving fast towards quantum computing with key figures like Pr. Julia Kempe (succeeding AI pioneer Pr. Yann LeCun at the head of the Center for Data Science at NYU since a few years ago).

Nevertheless more imminent and radical changes are already happening thanks to advanced electronics: e.g. hardware acceleration amplified by Amazon and Apple so-called “neural processors” (i.e. dedicated processors unlike general-purpose CPUs). In particular, low-power electronics is key as the non-storability of electricity is still an issue in spite of Apple, Tesla (and many others) research efforts.

#ai #deeplearning #electronics #quantumcomputing

I strongly recommend the reading of Marc Andreessen’s interview about Productivity and Scheduling.

It's a gold mine of information if you are looking to improve the way organize yourself and how you define your goals.

And let’s carry on the conversation at! Tell us how you organize your workflow and how you set your priorities.

#productivity #ceo #decisionmaker #workflow #process

“At most organizations, the bottleneck is at the top of the bottle.” — Peter Drucker

Hopefully, is here to help decision-makers ;-)

#ai #decisionmaker #productivity #workflow #ceo

Visiting last week my Ph.D. father Charles Bouveyron and my Ph.D. brother Pierre-Alexandre Mattei at INRIA Sophia-Antipolis for Ph.D defense rehearsal.

Thank you very much for the warm welcome to the whole Masaai team! Science at comes from hard work taught by this kind of inspiring people.

#ai #thankyou

The 10 years study CEO Genome Project* conducted on more than 2,000 CEOs concluded that one of the 4 main behaviors of successful CEOs was the ability to decide with speed and conviction. Good CEOs realize that a wrong decision may be better than no decision at all!

At, we help leaders make their decision with an AI-powered prioritization and smart analysis tool.

With, you hold the cards!

*Read full study in Harvard Business Review, May-June 2017 Edition

#decisionmaker #decisionmaking #ceo #csuite #managerialskills #productivity

Curated data collection is underrated. At, we know that information quantity is key for both humans and artificial intelligence algorithms but we also know that annotation quality is the crux of what is at stake:

  1. reliable information for enlightened decision-making only allowed by chosen and trustworthy sources for ethical and true data
  2. non-endogenous data (neither algorithms nor human beings should breathe their own smoke)
  3. relevant knowledge to protect your focus from social media distraction side effects.

“With great power comes great responsibility” like an uncle used to say, but without relevant information there is no responsibility and no power.

Surprisingly, a time delay of little more than 3 years can be observed between a dataset release and available AI products at human-level quality: handwritten text recognition in the 1990s, face recognition in the early 2000s, image and speech recognition in the early 2010s, automatic translation in the late 2010s.

In Natural Language Processing, Big Bertha-sizes datasets for OpenAI's GPT or UW Allen's Grover algorithms training confirms that trend paving the way towards new products.

At, this progress wave tells us that the time has come for better information tools for decision-making.

Hello everyone, I defend my Ph.D.


#research #phd #celebration

Today, the multiplication of information sources, collaborative tools and databases makes access to information increasingly difficult. Traditional research tools did not change much compared to the state of art in Research and the increasing volume of data.

The consequences for companies are extremely harmful: waste of time, focus disorder, loss of knowledge, increased stress, duplicate work, etc. Companies are well aware of the importance of knowledge accumulation but addresses the main problem of knowledge management: access to the right information at the right place when you need it.

#startup #ai #focusonwhatmatters