Cybersecurity and AI
Cybersecurity is a rapidly evolving domain with new, complex threats emerging daily. Security professionals deal with a large volume of alerts from a multitude of technologies, making threat detection very effort-intensive and people-dependent. To resolve these threats quickly, we can leverage AI capabilities and provide our experts with a powerful, intelligent support mechanism.
Introducing the team
The Atos MDR Data Science team is one of the driving forces behind Atos’s next-generation managed detection and response (MDR) service. The primary responsibility of this team is consolidating security use cases and translating them into threat detection algorithms (models) which are then integrated into AIsaac – Atos’s AI-driven MDR platform.
The team brings together highly skilled and qualified professionals from various domains to work together towards the common goal of solving complex cybersecurity problems.
apply AI techniques to cybersecurity and design models to detect advanced security threats
design and build the big data systems used by the data scientists. They are responsible for data acquisition, transformation and data system management
develop the AI models designed by data scientists and integrate the models into the AIsaac platform. They use Scala or Python to develop the AI models and provide a visualization layer to analyze the output
define the problem statement, design the cybersecurity algorithm, help data scientists with feature selection and validate the outcomes
This global team is spread across the US, Poland, France and India. Between them, they hold advanced degrees (Masters, PhD) and certifications in data science, analytics, software engineering, advanced computing and cybersecurity.
The team’s work
While data science and machine learning techniques have been successfully applied in multiple industries including healthcare, retail and marketing, this success is yet to be replicated on the same scale in cybersecurity. The primary challenges lie in the complexity and dynamic nature of threats.
The Atos MDR data science team leverages a variety of event and alert sources to build models that can not only detect existing security threats but also new threats that may emerge in the future. These models undergo a continuous refinement process so that the detection accuracy is high.
The team works at an extremely fast-paced, optimized tempo and has automated many of the steps in the model creation process. This ensures that the turnaround time for a new model (from ideation to production) is very fast and able to keep up with the changing security scenarios. While the models have been built to be plug-and-play elements for all customer environments, the modular approach we use for model development helps ensure that any required customization can be made easily and quickly.
The models developed by this team have shown success in spotting supply chain attacks, complex and latent malware, lateral movements, insider and targeted attacks.
The innovative work of the Atos data science team has led to creation of strong intellectual property in the form of patents (including US10270799B2 – Methods and systems for predicting vulnerability state of computer system and US10757122B2 – User behavior anomaly detection) and shaped an industry-leading MDR service. Atos continues to work on new threat detection scenarios at an aggressive pace, in order to help our customers stay one step ahead of attackers.
About the author
Global Head of Data Science, Managed Detection and Response (MDR), Atos
Harshvardhan currently heads the Data Science division for Managed Detection and Response (MDR) at Atos. His work involves establishing the vision, mission of using data science to detect advanced cybersecurity threats and overseeing the creation of various Artificial Intelligence (AI) models and algorithms used in AIsaac – Atos’s next-gen AI platform used for delivering MDR service.
Harshvardhan has been working in cybersecurity for 13 years, during which he has directly serviced large enterprises and Fortune 500 customers across US, Europe, and Asia Pacific. He currently holds 2 U.S. patents in AI & Cybersecurity. He is also a Certified Information Systems Auditor.
Interested in next publications?
Register to our newsletter and receive a notification when there are new articles.