JOINT SPECIAL COLLOQUIUM: AI-Optimized Advanced Packaging for Next-Generation Computing

JOINT SPECIAL COLLOQUIUM

College of Science, College of Engineering & Quantum Materials and Sensing Institute (QMSI)
AI-Optimized Advanced Packaging for Next-Generation Computing
Dr. Rabindra Das
MIT Lincoln Laboratory

Friday, Apr 3, 2026; 11:30am to 12:30pm
Hosts: Prof. Arun Bansil & Prof. Kin Chung Fong

Venue: Elliott Hall – Room 130C, 147 S. Bedford St, Burlington, MA
Remote: MS Teams Link

Abstract

The rapid growth of artificial intelligence (AI), high-performance computing (HPC), and data-intensive sensing systems is creating unprecedented demands for computational capability, energy efficiency, and system integration. Applications such as autonomous sensing platforms, satellites, and unmanned aerial and underwater vehicles increasingly require powerful onboard processing to analyze large volumes of data in real time. As conventional transistor scaling slows, advanced packaging and heterogeneous integration are emerging as critical technologies for enabling next-generation computing systems.

This talk presents a research vision for AI-optimized advanced packaging, where artificial intelligence techniques—particularly decision-tree-based optimization—are used to guide the design and fabrication of complex heterogeneous microsystems. AI-driven approaches enable optimization of substrate fabrication, chiplet placement, interconnect routing, power delivery, and thermal management across multi-chip systems. A central focus is the development of heterogeneous System-on-Wafer (SoW) architectures, integrating tens to hundreds of chiplets on a single wafer substrate to achieve extraordinary computing density. A case study on superconducting wafer-scale multi-chip modules with ultra-fine-pitch micro-bump interconnects demonstrates how advanced packaging can address key challenges in scalability, interconnect density, and system performance for future AI, HPC, and quantum computing platforms.

Biography

Rabindra N. Das, Ph.D. is a Member of the Technical Staff in the Advanced Technology Division at MIT Lincoln Laboratory, Lexington, MA. Previously, he served as a Principal Engineer at Endicott Interconnect Technologies (formerly IBM Endicott). Dr. Das has more than 23 years of experience in microelectronics packaging and heterogeneous integration, spanning high-performance computing, medical electronics, and superconducting quantum hardware systems. He has authored 135+ technical publications and holds 51 patents in microelectronics packaging technologies. He has been recognized for four consecutive years (2020–2023) in Stanford University’s list of the world’s top 2% most-cited scientists.