During my time as a Design Engineering PhD Research Intern at AMD, I built a Large Language Model (LLM) pipeline designed to automate hardware-aware GPU kernel generation for emerging architectures. I trained per-architecture open-weights models using multi-turn self-distillation to generate highly performant HIP and CUDA kernels. To support this pipeline, I automated kernel profiling and performance analysis, ultimately creating tools that accelerated the workflow for Application Engineers handling RFPs, system acceptance, and product launches.

As an Embedded Firmware Co-op on the Robotic Safety Systems team at Amazon Robotics, I integrated the company's robotic fleet with AWS IoT for real-time MQTT communication and connectivity monitoring. To streamline metric collection across all devices, I built a robust MQTT Metric AWS Cloud Development Kit (CDK). Beyond cloud integration, I debugged production hardware to resolve USB protocol failures, refactored Qt Quick QML code for improved modularity, and overhauled the service mode touchscreen interface to create a more intuitive experience for hardware deployed in fulfillment centers.

As a High-Performance Computing Software Validation Intern at AMD, I developed Python scripts and HIP/C/C++ kernel code to automate health metric collection for GPU systems within a SLURM cluster environment. I integrated AI-driven statistical analysis tools like Ollama and LlamaIndex to detect anomalies in these metrics, and I automated the reporting process using Prometheus alerts sent directly to Microsoft Teams. Additionally, I enhanced the reliability of our validation workflows by expanding testing frameworks with CMake, CTest, and CDash, and by building Grafana dashboards to monitor overall GPU node performance.

While working as a Co-op Research Assistant in the Cyber System Assessments group at MIT Lincoln Laboratory, I focused on the reverse engineering and dynamic analysis of embedded Linux systems across MIPS, ARM, and x86 architectures. Using tools like Ghidra, I disassembled shared objects and executables to pinpoint and resolve boot-process emulation failures. I successfully emulated devices using QEMU and PANDA, refined platform usability with custom bash scripts, and utilized Burp Suite to reverse-engineer network traffic, bypassing failed HTTP requests to restore full device functionality.

As a graduate Teaching Fellow at Harvard University for the “Computing at Scale” course, I support a class of 111 students. To facilitate their learning and evaluate their progress, I author problem sets and quizzes, and I hold weekly office hours throughout the term to help students work through complex concepts.

During my time as an Engineering Tutor for the IEEE-HKN Honor Society at Northeastern University, I tutored students in core electrical and computer engineering courses, including Embedded Design, Circuits & Signals, Engineering Algorithms, and Computer Architecture. I focused on simplifying complex topics through both one-on-one and group tutoring sessions to foster a deeper mastery of the material. Beyond academic help, I also leveraged my own experiences to advise new students on navigating their coursework, securing co-op opportunities, and planning their career paths.

A full-day workshop on agentic AI for computing systems design, spanning computer architecture, programming languages, and operating systems, held at ISCA 2026. Organized with Zishen Wan, Chenyu Wang, Andy Cheng, Shvetank Prakash, Arya Tschand, Ankita Nayak, and Prof. Vijay Janapa Reddi. The programme pairs keynotes and technical sessions with the launch of Architecture 2.0: Agentic Design Loops for Computing System Synthesis.

During my time as Vice-President of the Institute of Electrical and Electronics Engineers (IEEE) chapter at Northeastern University, I coordinated and led weekly technical workshops covering topics like PCB design, soldering, Python, and 3D printing. I also facilitated weekly meetings featuring presentations from leading industry professionals to help connect students with the broader engineering community. Furthermore, I co-organized the 2023 IEEE Communications Society School Series in collaboration with Northeastern's Institute for the Wireless IoT.

Serving as an Executive Lab Manager at the Boston Leadership Institute, I oversaw the logistical and operational needs of a massive summer educational program catering to 2,372 students across five locations. I managed the budget and material ordering for 56 courses, while administrating a team of 53 teachers and 11 site managers to ensure seamless delivery. Beyond streamlining internal processes and managing toxic waste disposal, I also contributed directly to the classroom by teaching organ identification, suturing techniques, and dissections.