Skip to main content
  • UC Berkeley
  • College of Letters & Science
Berkeley University of California

URAP

Project Descriptions
Spring 2026

URAP Home Project Listings Application Contact

Energy-efficient and High-Density Memory Technologies for AI-Hardware

Asir Intisar Khan, Professor  
Electrical Engineering and Computer Science  

Applications for Spring 2026 are closed for this project.

Data-centric applications face growing latency and energy challenges due to the separation of logic and memory components in traditional computing architecture. Neuro-inspired computing, modeled on the brain’s energy-efficient neural networks, offers a promising solution by reducing off-chip memory access. However, existing memory technologies struggle to achieve gradual conductance changes and multi-state operation, which are essential for neuromorphic systems. Similarly, 3D logic-memory integration holds potential to reduce latency and power consumption, but requires low-voltage, logic-compatible memory. Addressing these challenges demands fundamental advances in memory materials and device innovation.

In this project, we will develop high-density, low-power memory technologies using engineered chalcogenide and ferroelectric materials. By addressing critical limitations in neuromorphic systems and enabling 3D integration, this thrust will drive advancements in energy-efficient memory for next-generation data-centric applications.

Role: One or more of the following: Materials deposition, characterization, device fabrication, and electrical characterization. Data analysis, present and co-author publication.

Learning Outcomes: (i) Hands-on experience with advanced memory materials (chalcogenides and ferroelectrics), including thin-film processing, and electrical characterization; (ii) Understanding of key memory metrics and trade-offs relevant to AI hardware (switching energy, latency, endurance, retention, variability, and multi-state operation); (iii) Exposure to neuromorphic and 3D integrated memory concepts, including gradual conductance tuning and system-level implications of memory–compute co-design; (iv) Training in experimental data analysis, scientific writing, and presentation of results, with opportunities to contribute to conference/journal publications.

Qualifications: Strong motivation and curiosity to work collaboratively in an interdisciplinary research environment.

Hours: to be negotiated

 Mathematical and Physical Sciences   Engineering, Design & Technologies

Return to Project List

Office of Undergraduate Interdisciplinary Studies, Undergraduate Division
College of Letters & Science, University of California, Berkeley
Accessibility   Nondiscrimination   Privacy Policy