Resume
AI/ML Researcher & Engineer
Profile
AI/ML Researcher and Engineer pursuing a Master's in AI and Machine Learning at Drexel University. Research spans Bayesian neural network initialization, mechanistic interpretability for LLM safety, and extended pretraining of BioMedBERT. Previous experience includes building enterprise RAG pipelines and deepfake systems at Ernst & Young, and backend infrastructure at Neuflo Solutions.
Education
M.S. Artificial Intelligence and Machine Learning
GPA: 4.0Expected June 2026
Drexel CCI Merit Scholarship Recipient
M.Sc. Computer Science - Data Analytics
GPA: 3.77Graduated March 2022
Best Outgoing Student - Rajagiri College of Social Sciences
B.Sc. Physics
GPA: 3.39Graduated June 2020
Experience
Researcher - Prof. Rezvaneh Rezapour (COOP)
CurrentOctober 2025 - Present
- Contributed to longitudinal study of self-stigma among people who use drugs on Reddit: benchmarked 7 LLMs (F1=0.856), engineered prompts for 10 stigma indicators, backward evaluation (macro-avg F1=0.864)
- Extended BioMedBERT pretraining on PubMed abstracts (2020-2024): resolved MongoDB indexing bottleneck, migrated JSON to Parquet (2x compression), robust BLURB evaluation (13 tasks, 10 seeds)
Research Assistant - Prof. Jake R. Williams
CurrentOctober 2024 - Present
- Extended Bayesian Initialization to convolutional architectures using unfold operations and product quantization (81% MNIST with 20% data, 65% CIFAR-10 without gradient descent)
- Conducted multi-month hyperparameter ablation for 1.5B-parameter LLM (SAFFU) with Bayesian Parameter Initialization
- Managed GPU server operations and built training monitoring visualizations
Academic Roles
June 2025 - Present
- Course Assistant: CS 613, CS 501, CS 503
- Lab Assistant: CI 101, CS 164
- Graduate Dean's Ambassador: Representing CCI at virtual events
Backend Developer
September 2023 - August 2024
- Delivered backend infrastructure for NEET coaching MVP using FastAPI, PostgreSQL, Docker, and Azure
- Curated and built comprehensive NEET question bank: collected questions, answers, and images in LaTeX format, manually verified each entry end-to-end
AI/ML Engineer (Associate)
July 2022 - August 2023
- Built enterprise document search using Azure Cognitive Services, Blob Storage, Form Recognizer, and GPT-3.5 via FastAPI
- Developed digital twin characters using LLMs and deepfake lip sync for metaverse onboarding
- Delivered LLM-powered chatbot and prototyped Stable Diffusion text-to-image POC
AI/ML Research Intern
November 2021 - May 2022
- Developed YOLOv3-based form-field detection models for scanned documents
- Conducted research on LayoutLM for transformer-based document understanding
Publications
"Longitudinal Analysis of Self-Stigma: A Cognitive-Affective-Behavioral Characterization of Reddit User Timelines"
Layla Bouzoubaa, Hyung Wook Choi, Milan Varghese
"Deep Learning-based Sentiment Analysis on COVID-19 News Videos"
Milan Varghese and Anoop V. S.
15th International Conference on Information Technology and Applications (ICITA-2021)
Awards & Honors
Drexel CCI Merit Scholarship
Fall 2024 - 2026
EY Spot Award
Q4 FY23
Best Outgoing Student
Rajagiri College, 2022
Skills
Machine Learning
Data & Cloud
AI Development Toolkit
Languages
Certifications
Azure AI Fundamentals (AI-900)
Microsoft
Azure Fundamentals (AZ-900)
Microsoft
Workshops & Speaking
NIT Calicut's Tathva 2022
Invited speaker - Full-day Machine Learning workshop with 200+ attendees
Python & ML Training
Trained students from India and UK in Python programming and ML fundamentals
"Digital Convergence" Tech Session
Organized session covering ML, Quantum Computing, Blockchain, and IoT