๐Ÿ† BONUS

๐ŸŽฎ Research Philosophy & Character Stats

๐ŸŽ€ I am skeptical of AI progress that scales everything except accountability. ๐ŸŽ€

My research sits at the intersection of three questions: how do we adapt powerful models without requiring unlimited resources? How do we evaluate them honestly, beyond what aggregate metrics are designed to hide? And who are we building them for? These questions have taken me from parameter-efficient fine-tuning to adversarial benchmarking to low-resource NLP โ€” and increasingly toward mechanistic interpretability, where the goal is understanding not just what models output, but what they actually represent.

Efficiency, interpretability, and inclusion are not separate research agendas for me. They are the same argument at different levels of abstraction.

๐ŸŽ“ Education

  • 2023 - present
    PhD in Computer Science
    Boston University, Boston, MA, USA
  • 2017 - 2021
    BSc. in Computer Science and Engineering
    Shahjalal University of Science and Technology, Sylhet, Bangladesh

๐Ÿ’ผ Experience

  • Spr.'25,'26
    Graduate Teaching Fellow
    Boston University, Boston, MA, USA
  • 2022 - 2023
    Research Developer
    Wikimedia Foundation, San Francisco, CA, USA
  • 2021 - 2022
    Machine Learning Engineer
    GigaTech Ltd., Dhaka, Bangladesh
  • 2022 - 2023
    Research Assistant
    BRAC University, Dhaka, Bangladesh
  • 2021 - 2023
    Research Affiliate
    Bengali.AI, Dhaka, Bangladesh

๐Ÿค– Research Interests

Primary
Parameter-efficient Finetuning Mechanistic Interpretability Adversarial Attacks Machine Unlearning
Secondary
Bias Mitigation & Fair Representations Robust Domain Adaptation Knowledge Infusion AI For Social Good & Scientific Discovery

๐Ÿงญ Service and Leadership

  • Service
    • Reviewer - NeurIPS, ICML, ICLR, CVPR, ECCV, ACL-ARR, IJCV
    • Mentor - HSA , Intl. ML Olympiad
    • Volunteer - Outreachy, BDOSN Ada Lovelace Celebration,
  • Leadership
    • Founding Member | SUST ACM Student Chapter
    • Judge | SUST Tech Fest DL Enigma, BDOSN NLP Hackathon
    • Organizing Committee | IEEE ICBSLP, 2019. SUST Tech Fest, 2020

๐Ÿ… Accolades

    • ๐Ÿฅ‡ Champion, BRACATHON 3.0 - Healthcare, 2019
    • ๐Ÿฅ‡ Champion, SUST Tech Fest - Hackathon, 2019
    • ๐Ÿ… 4th place, AI-Based Dhaka Traffic Detection Challenge, 2020
    • ๐Ÿ’ป Grace Hopper Scholarship Recipient, 2021
    • ๐Ÿฅˆ 2nd Runner-up, CVPR AI City Challenge Track-4, 2022
    • ๐Ÿฅˆ 2nd Runner-up, Microsoft Imagine Cup SEANM, 2022
    • ๐ŸŒ Outreachy Internship at the Wikimedia Foundation, 2022
    • ๐ŸŽ“ Dean's Fellowship, Boston University, 2023
    • ๐Ÿ’ซ CRA-WP Grad Cohort Scholarship, 2025
    • ๐ŸŒŸ CVPR WiCV Mentorship Grant, 2025

๐ŸŒ™ Non-academic Interests

HOZIER ๐“‡ข๐“†ธ Malai Chaa โ˜• Analog Horror ๐Ÿ“ฝ๏ธ Folk Horror ๐“Š๐“‹ผ๐“Š๐“‹ผ๐“Š Gothic Horror ๐Ÿฆ‡ Cosmic Horror ๐ŸŒŒ Ritual Horror ๐Ÿ•ฏ๏ธ Ambient Horror ๐ŸŒซ๏ธ Magical Realism โœจ Fall of my enemies โ˜ ๏ธ Comics ๐Ÿฆธโ€โ™‚๏ธ Anime productions by Mappa ๐ŸŽŒ Seas with warm water ๐ŸŒŠ Anything by Neil Gaiman and Alan Moore ๐Ÿ“š
"I, for one, welcome our interpretable and computationally modest overlords." โ–‹