I am a PhD student in Computer Science at the University of California, Irvine, advised by Prof. Michael Franz in the Secure Systems and Software Lab.
My research sits at the conjunction of compiler and the machine it targets. My primary focus right now is on mitigating silent data corruption in large scale datacenters - the kind of quiet but catastrophic errors caused by defective hardware or even random particle strikes. For that, I build LLVM transformation passes to insert checks into the program that let a program detect transient hardware faults in itself. This checking instructions ensure that a corrupted value never escape into memory, a branch, or a call. I test the resilience of the hardened program using fault injection experiments accross differnt ISAs such as x86, Arm32, etc. To conduct fault injection experiments, I built a fault injection framework from the open source gem5 simulator. Those hands-on results are what drive my new mitigation strategies. Previously, I worked on cross-architecture binary translation, lifting legacy x86 binaries to LLVM IR and recompiling them for AArch64.
Before starting my PhD journey, I was a Lecturer in Computer Science at United International University in Dhaka. I taught several core programming lanugage courses including object-oriented programming and pattern recognition. Prior to that I built the backend of a nationwide high school admission system Link to XI admission system for Bangladesh’s Ministry of Education, which processed applications from 1.3 million students. I completed my BSc in Computer Science and Engineering at CSE, BUET. I conducted my undergraduate thesis on reducing blockchain storage overhead via sharding the ledger under Dr. Muhammad Abdullah Adnan.
PhD in Computer Science
University of California, Irvine
MS in Computer Science, 2026
University of California, Irvine
BSc in Computer Science and Engineering, 2021
Bangladesh University of Engineering and Technology (BUET)
LLVM, gem5, GDB, SVF, control flow graphs
C, C++, Python, Java, x86 & ARM assembly, CUDA
Spring Boot, Flask, Node.js, SQL, REST, Keycloak
Model Context Protocol, Gemini, OpenAI API
Git, Docker, Jenkins CI/CD, JIRA
Register allocation, SSA dataflow analysis, BFT & Paxos consensus, sharding
Doctoral research in compilers, computer architecture, and systems reliability, advised by Prof. Michael Franz.
Engineered the backend of a nationwide high school admission system serving 1.3 million applicants across 4 education boards.
GC overhead limit exceeded OutOfMemoryError in the 1.3M-applicant ranking query, resolving it by replacing entity-based retrieval with the JDBC API.Graduate Teaching Assistant. Mentored 300+ students across 6 computer science courses, assisting instructors with grading and question design.
Lecturer. Courses taught: