Project Description:

Transmitting and processing information traditionally involves electronic circuits, which use electrons to do the job. The research in this project uses photons instead, which are individual units of light. These photons have some properties that make them ideal for handling information. They can move fast and carry a lot of data. However, there are still some challenges to overcome when using light for these circuits, like directing it where we want it to go, preventing it from interfering with other light, and switching it on and off when needed.

One of the materials used for making electronic circuits is silicon, which is inexpensive and can be mass-produced in large quantities. When it comes to PICs, however, there is a need to use different materials on top of silicon, which is not easy. In this project, we will seek to create new materials that will serve as the building blocks for the PICs of the future. These materials have special properties, including mutual compatibility, and improved control over light-matter interactions. Furthermore, these materials can be directly deposited and integrated on chips, unlike the current hybrid materials prepared by lithography and patterning methods, which can require additional processing steps.

We'll need help in simulating new photonic materials at the smaller scale (materials level), as well as the circuit level to help guide planned experiments.

Requirements:

Majors: We're particularly interested in ECE, MSE, and ME majors; all College of Engineering majors can apply

Requirements: Experience with programming in Python, C/C++, and/or MATLAB

Desired experience: Enthusiasm for scientific programming. Understanding of electromagnetism (e.g., it's helpful to have previously taken ECE 30411).

Academic Years Eligible: Juniors and seniors with the desired experience will be preferred, but all undergraduates are also eligible to apply.

Project Description:

This project will consider the formulation and solution of engineering applications (chemical superstructure optimization, optimal power flow) using quantum-enhanced methods. We will learn, through libraries in Python and Julia, how to address these problems and efficiently solve them.

Requirements:

Project Description:

The broad goal of this project is to explore the potential of machine learning in optimizing computer networks, and in designing computer networks for machine learning workloads.

The set of projects is broad and include:

1) Intent-driven networking: synthesizing network configurations (e.g., firewall rules) from policies specified in natural language

2) Automated network diagnosis: Automatically diagnosing network performance problems based on trouble tickets in natural language.

3) Designing simulators for studying real-world GPU scheduling algorithms. Such algorithms are important for distributed training of Large Language Models.

4) Designing low latency live 360 degree video streaming systems. Here, there are more systems oriented projects, as well as projects involving use of ML to drive the design of algorithms behind video streaming.

5) Leveraging programmable network hardware (e.g., programmable switches, FPGA-based NICs) for ML inference and training.

Based on the student's specific interests, goals and background, an appropriate project would be determined.

Requirements:

Project Description:

Autonomous experimentation (AE) represents a transformative leap in materials innovation, enabling rapid, data-driven discovery that far outpaces traditional design-of-experiment approaches. By combining robotic automation with AI-guided decision-making, it holds the potential to unlock novel materials and processes critical for advancing technologies in microelectronics, energy, and beyond.

This research opportunity focuses on accelerating materials innovation for microelectronics through AI-enabled AE. The project will involve the development or enhancement of a robotic experimental platform capable of high-throughput synthesis and in-line characterization of material systems relevant to next-generation electronic devices. Central to the effort is the creation of an adaptive, closed-loop experimental workflow powered by machine learning algorithms that guide experimental design in real time based on evolving results. The project will also involve the assembly of a FAIR (Findable, Accessible, Interoperable, Reusable) database cataloging experimental parameters, material compositions, process conditions, and measured properties. The project aims to demonstrate the accelerated discovery and optimization of material properties—such as dielectric constants, piezoelectric properties, or thermal stability—thereby reducing the time and cost traditionally required in microelectronics materials development.

Requirements:

This project requires someone hands-on in the lab who can build and code. It is preferred to have some knowledge of materials and material characterization. The project is truly cross-disciplinary. Background in any of electrical, mechanical, materials, robotics, AI is welcome!