Ahmad Farhan Shahriar Chowdhury

Professional Experience

BRAC University — Dhaka, Bangladesh June 2025 – Present

Lecturer (Full-Time), Department of Computer Science and Engineering

Courses: Operating Systems, Computer Architecture, Computer Graphics, Machine Learning

UMass Boston — Massachusetts, USA September 2025 – Present

Research Intern under Dr. Tarikul Islam Papon

Education

Bangladesh University of Engineering and Technology — Dhaka, Bangladesh June 2025 – Present

M.Sc. in Computer Science and Engineering

Bangladesh University of Engineering and Technology — Dhaka, Bangladesh February 2020 – March 2025

B.Sc. in Computer Science and Engineering CGPA: 3.88 / 4.00

Relevant Coursework: Structured Programming Language, Data Structures and Algorithms, Compiler, Computer Architecture, Operating Systems, Computer Security, Computer Graphics, Machine Learning

Publications

ZARC: Re-inventing Zone Allocation for LSMs on ZNS SSD

Farhan S. Chowdhury, Shadman Saqib Eusuf, Subhadeep Sarkar, Tarikul Islam Papon

Proceedings of the International Workshop on Data Management on New Hardware (DaMoN), 2026

Under Review PLAZA: Proximity & Lifetime-Aware Zone Allocation for LSM Engines on ZNS SSDs

Farhan S. Chowdhury, Shadman Saqib Eusuf, Subhadeep Sarkar, Tarikul Islam Papon

VLDB 2026

Research Projects

Extending Support for RocksDB on Zoned Namespace SSD August 2025 – Present

Advisor: Dr. Tarikul Islam Papon, UMass Boston

Currently working on extending ZenFS by implementing different algorithms and policies:

New Zone Allocation: Deciding which zone to select for writes when a new zone is needed.
File Placement: Deciding where to put a new SST file based on its properties.
Garbage Collection: Deciding which files to move in order to facilitate zone erasure.

Memory-Efficient ML-based Prefetch Filtering (UG Thesis) February 2024 – March 2025

Advisor: Dr. Rifat Shahriyar, BUET

Developed a Neural Network-based Prefetch Filtering algorithm and a training & testing pipeline in the ChampSim cache simulator; compared with the existing Perceptron-based filter (PPF).
Achieved a reduction of 93% in model weights and 47% in total memory overhead while keeping performance almost the same as the reference model.
Significantly improved Coverage compared to the baseline prefetcher without any significant loss of Accuracy.

ATLAS: Automated Tree-based Language Analysis System for C and C++ source programs October – November 2025

Advisor: Dr. Mahmuda Naznin, BUET

Developed a Static Analysis Tool for C and C++ that generates interconnected Control-Flow and Data-Flow Graph visualizations.
Handled problematic memory-related features such as control-flow through function pointers and tracking variable changes through indirect pointer access.

Selected Academic Projects

Computer Architecture (3-1) GitHub
Created a 4-bit computer from scratch using basic ICs that can run MIPS instructions, along with an assembler and LogiSim/Proteus simulations.
Compiler (3-1) GitHub
Implementation of a basic C compiler in C++, Flex, and GNU Bison that compiles C code to 8086 assembly, featuring Lexical Analysis, Syntax Tree Generation, and Peephole Optimization.
Operating Systems (3-2) GitHub
Implementation of Lottery Scheduling, Multi-Level Feedback Queue Scheduling, Multithreading, Mutex Locks, and Semaphores in the Xv6-riscv kernel (C).
Computer Networks (3-2) GitHub
An FTP server from scratch using Socket Programming in Java; TCP Adaptive Reno Congestion Control Algorithm in ns-3.
Computer Graphics (4-1) GitHub
Implementation of Ray Tracing using C++ and OpenGL.
Machine Learning (4-2) GitHub
Classification via Logistic Regression with Ensemble Learning (Stacking & Bagging); a Feed-forward Neural Network with Batch Normalization, Dropout Regularization, and Adam Optimization — all from scratch in Python.

Skills

Languages:

Bash, C, C++, Java, JavaScript, RISC-V Assembly, Python, SQL, x86 Assembly

Research Tools:

ChampSim, Scikit-learn, TensorFlow

Dev Tools & Frameworks:

Docker, Express.js, JavaFX, Node.js, MongoDB, PostgreSQL, RocksDB

Other Technologies:

ATmega32, Arduino, Flex, GNU Bison, OpenGL, Proteus, Xv6

Ahmad Farhan Shahriar Chowdhury

Research Interests

Professional Experience

Education

Awards & Achievements

Publications

Posters & Presentations

Research Projects

Selected Academic Projects

Skills

References