Senior Research Scientist @ IBM Research

Innovating at the intersection of AI and Security

Leading Large Language Models Customization for complex domains.

Abdulhamid Adebayo

Biography

Dr. Abdulhamid Adebayo is a Senior Research Scientist at IBM’s T.J. Watson Research Center. His expertise lies at the critical juncture of Data Engineering, Cybersecurity, and AI Customization. Currently, he leads the Data Processing and Operations team, where he spearheads the development of scalable pipelines that transform raw data into the high-quality tokens required for foundational Large Language Models (LLMs).

He earned his Ph.D. in Computer Science from Howard University, where his doctoral research focused on Secure Dynamic Spectrum Access and Wireless Network Virtualization. During his time at Howard's Data Science and Cybersecurity Center (DSC2), he pioneered security frameworks for 5G and beyond, establishing a foundation in robust, adversary-resistant system design that he now applies to AI infrastructures.

At IBM, Dr. Adebayo has successfully bridged the gap between academic theory and enterprise application. His work has resulted in multiple patents and high-impact open-source contributions, most notably the IBM Data Prep Kit, which democratizes access to large-scale, high-fidelity data processing for the global research community.

Open Source Impact

Core Contributor

IBM Data Prep Kit

A community-led toolkit for scaling unstructured data preparation for LLMs. Enables high-quality token generation across clusters with thousands of CPU cores.

PythonRaySparkKubernetes

Patents & IP

US Patent 12,468,856

AI-assisted compliance mapping

Methods for automating the mapping of regulatory requirements to technical security controls using machine learning.

US Patent 12,413,557

Trusted execution environment for service mesh

Secure communication frameworks for cloud-native applications using Zero Trust principles.

US Patent 11,625,272

Scalable operators for automatic management of workloads

Hybrid cloud management systems for dynamic workload orchestration.

Selected Research

Initialize Contact

Have a question about the Data Prep Kit or interested in LLM/SLM collaboration? Drop a message into the terminal.

Accepting Connections