Amazon Traffic Engineering

• Designed a TrustZone AFL syscall fuzzer for the secure world in ARM Trusted Execution Environment (TEE).

• Built custom Trusted Applications in C, for OPTEE-OS and tested buffer overflow vulnerabilities using AddressSanitizer.

• Analyzed the kernel image binaries using Angr.io and Ghidra, and detected 2 overflow vulnerabilities using these tools.

• Contributed to OPTEE-OS core by patching handlers for Undefined Behavior Sanitizer, and 7 other kernel panic bugs.

• Configured Syzkaller implementation for multiple operating systems.

Project Mentor:

Prof. Dongyoon Lee

• Teaching Assistant for graduate-level Operating Systems (CSE506).

Teaching Assistant for the Fall 2023 semester for CSE-312 -Ethical, Social and Legal Issues in Information Systems, under Prof. Gino Elia.

As part my Advanced Project, I have been working on an iOS application - EyeCanDo - which operates completely on eye movement. The app is specifically for ALS patients with limited muscle movements. My contributions include:

• Upgrading the eye tracker calibration using homography with the help of OpenCV in C++.

• Developing a Phrase-level Gesture Typing ML model using BERT base model for transfer learning.

• Integrating the calibration module as well as ML model deployment on iOS using Swift.

Project Mentors:

• Prof. Xiaojun Bi

• Prof. Fusheng Wang

• Rui Liu

• Maozheng Zhao

For more information about the project: http://eyecando.org/