Bora Kizil

CEO

A computer engineer with an MBA from HEC Paris, Bora has been an entrepreneur since 2011. He joined Ezako in 2017 to accelerate the company growth.

Written by Bora Kizil

Three underwater platforms at three depths, each with its own signal, linked to one shared model.

Ezako and XPERT: deep learning for every underwater platform

Ezako and XPERT start a development project to adapt neural networks to every underwater platform, from drones to hydrophones, and to improve the real-time detection of underwater anomalies.

A series with one event labeled by hand and two more of the same shape found, outlined in dashes.

Upalgo Labeling joins the Altair Partner Alliance

Altair distributes Upalgo Labeling through the Altair Partner Alliance: a tool to annotate time series and sensor data, with label propagation.

The peaks of a spectrum, with the one that stands out boxed.

VASCREEN: detecting dangerous and illegal substances in luggage

Within the SecurIT programme, Ezako and MION develop VASCREEN, a solution to detect potentially dangerous and illegal substances in luggage and goods.

Three sensor series of different tempers, one event boxed in each.

What can you do with time-series data?

What time-series data is, where it comes from, how machine learning puts it to work, and what return to expect from it.

The same node, larger at each step of a rising line.

Ezako joins the Minc Scale program

Ezako joins Minc Scale, a programme that supports growing companies with international ambitions, to approach the Nordic market.

A field of computing cores, with a wave of work crossing it.

Ezako joins NVIDIA Inception

Ezako joins NVIDIA Inception, the programme that supports startups working with artificial intelligence.

A timeline of a machine’s running stretches, ended by three failures, with the span between them measured.

What is MTBF, and why does it matter?

The mean time between failures says how reliable a machine is: what it is, why it matters, how to calculate it and how to improve it.

A spectrogram with two sources boxed: a whistle and a train of harmonics.

Ezako and Alseamar: deep learning on autonomous underwater platforms

Within the RAPID project DASMIA, Ezako and Alseamar develop deep learning on autonomous underwater platforms.

Three series kept on three sites, each with a labeled event, joined to one shared model.

What is federated learning?

Federated learning trains a shared model on data that stays on the devices that hold it: only the model’s updates travel. What it is, and when it is useful.

The spectrogram of an engine run-up, its orders rising, with the abnormal stretch that cuts them boxed.

A deep learning approach to signal anomaly detection, with Safran Aircraft Engines

Abnormal observations in high-frequency sensor series can make the spectrograms of an engine test unusable. A deep learning approach to detect them, developed with Safran Aircraft Engines.

A sensor series inside its normal range, with one excursion out of it boxed.

Why anomaly detection matters to companies

Anomaly detection finds the unusual patterns in a company’s data before they become serious problems: what it is used for, and why it matters.

A sensor series with two anomalies, each boxed and labeled.

What is supervised anomaly detection?

Supervised anomaly detection trains a model on labeled data, where unsupervised methods look for anomalies on their own: how it works, and what it takes.