AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Principal Applied Scientist, Multimodal GenAI @Oracle/OCI | Ex Microsoft GenAI | AI Specialist | LLMs & Multimodal Models Expert | Building Next-Gen Language & Vision Systems | Experienced in ML, NLP & Deep Learning
I’m an AI research engineer specializing in Large Language Models (LLMs), multimodal systems, and end-to-end GenAI pipelines. My work focuses on training, fine-tuning, and optimizing models across modalities -- text, image, video -- driving innovations from prototype to production in products like Microsoft Designer and Seeing AI. My work includes training and fine-tuning models using techniques like LoRA to dramatically reduce training time and data requirements for image generation. I’ve built evaluation frameworks that leverage LLMs (e.g., GPT-4) to automatically assess prompt adherence, output quality, and correctness—replacing weeks of manual effort. I’ve also customized and deployed image transformation and restyling and inpainting systems now integrated into creative design tools. I am the core contributor of MM-ReAct, an open-source agentic multimodal reasoning framework that combines LLMs with domain-specific tools for visual math, spatial reasoning, and multi-image understanding. I led the development of MM-Vid, a system designed for deep video understanding using GPT-4V and custom video/audio tools. MM-Vid enables long-form video reasoning, speaker identification, and GUI co-piloting, and has been successfully integrated into real-world accessibility tools. Previously, I worked on product recommendation, multilingual demography prediction, sentiment analysis, and video retrieval -- designing and building and deploying models using transformer-based architectures (e.g., XLMRoBERTa, CLIP4Clip). As an Adjunct Faculty at the University of Washington, I teach core computer science courses and mentor students in applied AI. I hold a PhD in Computer Science from Stony Brook University, where I focused on NLP, accessibility, and intelligent user interfaces.
Stony Brook University
MS, Computer Science
January 1, 2008 – January 1, 2011
Stony Brook University
PhD, Computer Science
January 1, 2008 – January 1, 2012
Bangladesh University of Engineering and Technology
BSc, Computer Science & Engineering
January 1, 2000 – January 1, 2006
Oracle
Principal Applied Scientist, Multimodal GenAI, OCI/Oracle
July 1, 2025 – Present
University of Washington
Part Time Faculty
January 1, 2023 – Present
Microsoft
Senior Research Software Engineer at Microsoft Gen AI
September 1, 2022 – January 1, 2025
Microsoft
Senior Research Software Engineer at Multimodal AI, Cognitive Services
October 1, 2018 – September 1, 2022
Microsoft
Research Software Engineer II at Deep Learning Team, Microsoft Research
August 1, 2015 – September 1, 2018
Microsoft
Software Engineer II at CSV, Microsoft Commmerce
December 1, 2013 – July 1, 2015
International Cross-Disciplinary Conference on Web Accessibility (W4A) 2013
Conference Reviewer
February 1, 2013 – March 1, 2013
Rio de Janeiro, Brazil
NetApp
Software Engineer, Storage Management System
August 1, 2012 – November 1, 2013
Morgan Stanley Smith Barney
Technology Analyst
June 1, 2011 – August 1, 2011
New York, New York
Charmtech LLC
Research Assistant
January 1, 2011 – July 1, 2012
Stony Brook, New York
Morgan Stanley
Technology Analyst
May 1, 2010 – August 1, 2010
Brooklyn, New York
Microsoft Research
Research Intern
June 1, 2009 – August 1, 2009
State University of New York at Stony Brook
Guest lectures
April 1, 2009 – May 1, 2009
Stony Brook, New York
State University of New York at Stony Brook
Teaching Assistant
January 1, 2008 – May 1, 2009
Stony Brook, New York
State University of New York at Stony Brook
Research Assistant
January 1, 2008 – July 1, 2012
Stony Brook, New York
Commlink Infotech
Member R&D
January 1, 2007 – December 1, 2007
Dhaka, Bangladesh
Streamstech Inc.
Member, Design and Development
August 1, 2006 – December 1, 2007
Chantilly, Virginia
Cultural Fit Analysis
The candidate demonstrates a strong cultural fit for a research-heavy, innovative environment, given their extensive background in Microsoft Research and their contributions to advanced AI projects. Their experience as a part-time faculty member also suggests a willingness to share knowledge and contribute to a learning culture. The diversity of projects, from core ML model development to system integration and optimization, indicates adaptability and a broad interest in the ML domain.
Soft Skills & Operational Fit
The candidate's extensive experience at Microsoft in various research and engineering roles, coupled with part-time faculty experience, suggests strong problem-solving, collaboration, and communication skills. Their work on optimizing models and building production-ready systems indicates a practical, results-oriented approach. The academic background and teaching experience also point to an ability to articulate complex concepts and potentially mentor others.