AiThority Interview with Dr. Arun Gururajan, Vice President, Research & Data Science, NetApp

Dr. Arun Gururajan, Vice President, Research & Data Science, NetApp highlights advancements in AI-driven ransomware protection, data management, and the importance of continuous learning, data quality, and ethical practices in AI development.

_______

Hi Arun, welcome to our AiThority interview series. Please share your journey and experience being in the AL/ML space? And key learnings?

I am currently the VP of research and data science at NetApp, where I lead AI/ML efforts across NetApp’s product organization. With my master’s and Ph.D. in computer vision & machine learning, I have been in the field of data science for more than 20 years and consider myself fortunate to be part of the amazing transformation that has happened in the field of AI. Over the years, I have led diverse teams across top tech companies and spearheaded numerous AI initiatives, including developing anti-phishing models, search & recommendation systems, conversational AI, fraud detection, and ransomware detection to name a few.

A key learning from my experience is the importance of continuous learning and adaptation. This has been very important in my ability to stay at the cutting edge of the rapidly evolving field of AI/ML. The other thing that I really value as a practitioner is to build a strong data foundation. High quality data is the backbone of any AI system. To that end, investing effort in creating robust data pipelines with solid data governance processes ensures that models are trained on high-quality datasets, ensuring reliable outcomes.

Also Read: AiThority Interview with Carolyn Duby, Field CTO and Cyber Security GTM Lead at Cloudera

Can you talk about some of the recent AI-driven innovations at NetApp that you find particularly exciting?

At NetApp, we are pushing the boundaries of AI in a couple of different ways. We are leveraging ways to use AI/ML to improve our offerings for customers and innovating data storage and management technology to make it easier for customers to use their data in AI applications.

One of the ways NetApp is using AI to help customers is the recently announced Autonomous Ransomware Protection with Artificial Intelligence (ARP/AI) offering which provides real-time detection and response capabilities that minimize the impact of cybersecurity threats. With this offering, NetApp is the first storage vendor to offer AI-driven, on-box ransomware detection with externally validated top-notch protection effectiveness. SE Labs, an independent cybersecurity testing laboratory, has validated the effectiveness of ARP/AI with a AAA rating after testing demonstrated a 99 percent detection rate for advanced ransomware attacks without flagging any false positives. Storage is the last line of defense against cybersecurity attacks intended to compromise an organization’s data and so secure storage is an important part of a complete defense-in-depth cybersecurity strategy. Using AI to identify and mitigate cyberattacks before they impact the organization’s data allows IT teams to rest assured their data is protected while reducing the operational burden and skills required to maintain their intelligent data infrastructure. Using AI also allows the software to continuously evolve and adapt to new threats, ensuring that NetApp ARP/AI remains at the forefront of protection effectiveness.

NetApp also recently made its AI-driven NetApp BlueXP classification service available as a core BlueXP capability, available to customers at no additional cost. This unique capability gives NetApp customers the ability to automatically classify, categorize, and tag data across their entire data estate to deepen data intelligence, enhancing efforts in governance, security, and compliance while enabling strategic workloads. This helps organizations better understand, govern, and leverage their data, especially as they apply it to GenAI and retrieval augmented generation (RAG) innovation.

In terms of enabling AI innovation for customers, NetApp offers advanced, all-flash storage that can serve data to high-performance technology stacks running AI workloads. NetApp also works closely with the major hyperscaler cloud providers to offer GenAI toolkits that allow organizations to use foundation models (FMs) for RAG processes. This enables organizations to apply proprietary data to unlock specific and relevant insights without exposing private data to a larger model.

Also Read: Role of AI in Cybersecurity: Protecting Digital Assets From Cybercrime

What are some of the biggest challenges NetApp faces in implementing AI solutions within its diverse infrastructure and how does NetApp address these challenges?

The key blockers for deployment of AI are concerns over privacy, governance and risk. Data is the foundation of AI and organizations want to tightly safeguard their proprietary data in order to maintain their competitive advantage. In addition, in a hybrid cloud world, datasets are distributed between on-premises and the cloud, which brings in another problem – that of data mobility.

NetApp’s robust data management solutions provide seamless data movement across on-premises and cloud environments, essential for a hybrid world where data needs to be agile and accessible. This capability ensures that AI models can access and process data wherever it resides without compromising performance or security. With its intelligent data services, NetApp provides organizations the capability to curate the data that is used to form the knowledge base for a RAG application. This essentially ensures that the application does not leak an organization’s confidential data. By providing secure and compliant data management solutions, NetApp empowers enterprises to accelerate their AI initiatives, enabling innovation and operational efficiency in a hybrid IT landscape.

Based on your current role, what emerging trends in AI and data science do you see as most pivotal for the future of enterprise IT infrastructure?

AI-driven automation is revolutionizing IT operations, leading to the development of self-managing systems that minimize human intervention and maximize efficiency and resulting in significant cost savings and streamlined processes. A good example is AIOps, which leverages machine learning to analyse large volumes of data generated by IT systems to identify patterns and anomalies that help in predictive maintenance. As an example, NetApp leverages AIOps to monitor the health of our storage systems deployed at customer locations and proactively ship replacements when we detect an increased risk of failure, thus avoiding disruptions to our customers.

The second emerging trend that I foresee is a heightened focus on security for AI. With the advent of generative AI, the attack surface for bad actors has significantly increased. To counter the slew of emerging attacks, organizations will start to bridge the gap between cybersecurity and data science teams. This will lead to the emergence of new data-driven auditing paradigms for algorithmic risk such as “Large Language Model (LLM) Red Teaming”, where the robustness of a system is tested via an adversarial approach. These approaches will be particularly honed to extract undesirable responses from a model.

The final and related trend I would mention is that of data privacy. For many organizations that are using AI to understand data about their customers, there are associated implications for the exposure of personal data. However, a higher degree of privacy (such as complete encryption) reduces data utility because it essentially renders the data unusable for a data persona. I foresee the emergence of tools that balance data privacy and data utility based on regulatory requirements.

Also Read: AI and Social Media: What Should Social Media Users Understand About Algorithms?

With AI increasingly integrated into core business processes, how does NetApp ensure ethical practices and governance in AI development and deployment?

 One of the most critical parts of ethical AI is maintaining the privacy and integrity of the data used for AI applications. In addition to enabling secure and programmatic RAG processes, NetApp tackles this issue by providing what we believe is the most secure storage on the planet. In addition to ARP/AI, NetApp also has capabilities to create immutable copies of data that ensure that malicious actors cannot tamper with the data to manipulate the outputs of an AI model.

Also, NetApp BlueXP classification empowers organizations to seamlessly identify and tag sensitive data across their entire operating environment enabling secure and compliant AI/ML pipelines via effective data curation.

How does NetApp leverage AI to enhance cloud operations and data services for clients managing hybrid multi-cloud environments?

With hybrid cloud becoming a central strategy rather than a transitional phase, NetApp is helping customers by providing cost optimization and predictability, standardized data management, and operational efficiency. Consistent data management across diverse environments is essential, and AI supports this by providing tools for data unification, ensuring that data security, disaster recovery, and provisioning processes are standardized across all environments.

The NetApp BlueXP platform provides unified control over storage and data services across hybrid multi-cloud setups, integrating powerful AIops capabilities to streamline operations. NetApp also integrates advanced data protection and cybersecurity measures to safeguard data integrity and enhance our customer’s resilience against threats so they can focus on leveraging that data to unlock business insights.

NetApp collaborates with key players across the technology space including the major hyperscale cloud providers and technology innovators like NVIDIA, Lenovo, and Cisco to develop converged infrastructure solutions that simplify storage and data management. These collaborations enable the development of comprehensive, enterprise-grade capabilities that can operate advanced workloads across hybrid multi-cloud environments for AI and beyond.

Finally, what advice would you give to professionals looking to build a career in AI and data science, especially in the enterprise tech industry?

For professionals looking to build a career in AI and data science, especially in the enterprise tech industry, my advice would be to continuously build a strong foundation in both theoretical and practical aspects of AI and data science. Staying up to date with the latest trends/advancements, focusing on developing soft skills, connecting with experienced professionals for guidance and building a strong professional network are all crucial components for breaking into a data science role. Additionally, keep looking for opportunities to work on real-world problems and projects, as hands-on experience is invaluable.

Thank you, Arun, for your insights; we hope to see you back on AiThority.com soon.

[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]