Back to portfolio


Incorporated startups

See where things are not running smoothly

profilePhoto   Numa Schenone
Linksium Contact Numa Schenone +33 (0)7 78 09 11 94


  • Increased availability of machines
  • Reduced maintenance costs
  • Rapid prototyping of decision algorithms

Key words

  • Predictive maintenance
  • Machine Learning
  • Artificial Intelligence

Intellectual Property

  • 5 software

Partnerships & Rewards

  • 2019 French Deeptech Innovation Competition Winner
  • 2021 i-Nov French Deeptech Entreprise Competition Winner




  • CNRS
  • UGA

Linksium Continuum

  • Maturation
  • Incubation
  • Acceleration


  • Incorporated startups


The maintenance of machinery is usually based on corrective and preventive methods that have the disadvantage of overestimating the need for inspections and shutdowns without minimising the risk of breakdowns.
There is currently an underlying trend in the world of industry 4.0, that of predictive maintenance using multiple data from different sensors to predict faults, ageing and end of life in a piece of equipment. The technology of Amiral Technologies, based on an innovation of the laboratory at CNRS GIPSA-Lab, is aimed at equipment manufacturers and providers of industrial IoT platforms.


Amiral Technologies offers:

  • A unique innovation in algorithms for automatic feature generation, regardless of the nature of the input time signals (electric signal, temperature, humidity, vibration, etc., or a combination of these signals) and without the need for professional expertise. The features generated in this way are richer and more discriminating, and produce higher performance with the AI and Machine Learning algorithms.
  • Generic and specific AI models, ready to be adapted to the specific requirements of the equipment in order to interpret, predict and generate relevant maintenance alerts: faulty equipment, signs of ageing and the approach of the end of working life.


  • Express-design predictive models (a few days rather than several months)
  • Immediate savings in maintenance costs (30 to 50% depending on the current process)
  • Immediate improvement in equipment availability
  • Up to 100% (results obtained on a real industrial dataset)

State of progress

The technology was implemented at the laboratory and validated on industrial data.
Amiral Technologies won the Digital Industry Program challenge organised by General Electric for predictive maintenance for their client HP.
A number of industrial studies are in progress to obtain a full proof of concept on sites.


  • Prediction of the duration of working life: Generic model and its application for industrial contactors and turbo reactors.
  • Prediction of defects: Generic model and its application to railway points and industrial printers.
  • Prediction of ageing: Generic model and its application to wind turbines and induction motors.

Featured articles

Nora Dempsey and deeptech innovations

1, 2 and 3 startups for Nora Dempsey, National Medal of Innovation 2021

Nora Dempsey is one of the four winners of National Medal for Innovation 2021 awarded by CNRS. Nora Dempsey, born in Ireland, has gained an international recognition for her research in Grenoble on...

Read the article
Dd amiral 2018 Frenchweb

Amiral Technologies, la startup IA grenobloise spécialisée dans les solutions de maintenance prédictive, intègre Station F

An ambitious promise: to invent a zero-defect and zero-downtime future for industry. Amiral Technologies, a startup co-founded and boosted by Linksium, based on the technological prowess of the Gre...

Read the article