Understanding Transformers and Attention Mechanisms
A deep dive into the transformer architecture and self-attention mechanism that powers modern NLP models.
#Transformers#NLP#Deep Learning
Read more
Welcome
Hi, I'm Milan Varghese
AI/ML Researcher & Engineer
Master's student in AI and Machine Learning at Drexel University (GPA: 4.00). Currently working as a Research Assistant developing novel neural network initialization algorithms and optimizing large language models.
About Me
Current Focus
Bayesian Neural Network Initialization
Developing novel initialization algorithms achieving 81% on MNIST with 20% data
PyTorch Neural Networks
Semantic Shift Analysis of PubMedBERT
Yearly continual pretraining on longitudinal PubMed corpus with embedding-space and concept-proximity analyses to quantify evolving biomedical knowledge
NLP BERT
I'm an AI/ML Researcher and Engineer pursuing a Master's in AI and Machine Learning at Drexel University. My research focuses on neural network initialization and semantic shift analysis in biomedical NLP. Previous experience includes building enterprise RAG pipelines and deepfake systems at Ernst & Young, and backend infrastructure at Neuflo Solutions.
Expertise
Tech Stack
Languages
Python Java C++ R
ML/DL Frameworks
PyTorch TensorFlow Keras Scikit-learn LangChain
Cloud & Databases
Azure AWS GCP PostgreSQL Oracle SQL
Tools & Platforms
Docker FastAPI Flask Git Power BI Tableau
Certifications
Azure AI Fundamentals (AI-900)
Azure Fundamentals (AZ-900)
Neural Networks and Deep Learning
Machine Learning (Stanford)
Background
Experience & Education
🎓 Education
M.S. Artificial Intelligence and Machine Learning
Drexel University, Philadelphia, PA
Expected May 2026 | Drexel CCI Merit Scholarship Recipient
M.Sc. Computer Science - Data Analytics
Mahatma Gandhi University, India
Graduated 2022 | Best Outgoing Student
💼 Work Experience
Researcher - Prof. Rezvaneh Rezapour (COOP)
Drexel University
Current Oct 2025 - Present
- • Longitudinal analysis of internalized self-stigma among Reddit users
- • Building GPT API pipelines for social media analytics
- • Fine-tuning BioBERT for medical text semantic shift detection
Research Assistant - Dr. Jake R. Williams
Drexel University
Current Oct 2024 - Present
- • Developed novel Bayesian Initialization algorithm (81% on MNIST with 20% data)
- • Analyzing semantic shift in PubMedBERT via yearly continual pretraining on longitudinal PubMed corpus
- • Configured compute servers for high-throughput training
Associate - Backend Developer
Neuflo Solutions Pvt Ltd
Aug 2023 - Aug 2024
- • Built backend infrastructure using Azure, PostgreSQL, FastAPI, Docker
- • Led intern training programs and client acquisition strategies
AI/ML Automation Engineer
Ernst & Young
Jul 2022 - Aug 2023
- • Built LLM-driven digital-twin modules and deepfake systems
- • Designed RAG pipelines on Azure with GPT-3.5
- • Developed Stable Diffusion POCs and LLM-powered chatbots
AI/ML Research Intern
Zoho Technologies
Nov 2021 - May 2022
- • Prototyped YOLOv3-based models for form-field detection
- • Researched transformer-based document understanding (LayoutLM)
Portfolio
Featured Projects
🔬 Research Projects
Bayesian Neural Network Initialization
ResearchNovel initialization algorithm using co-occurrence statistics and product quantization. Achieved 81% accuracy on MNIST with 20% data, 65% on CIFAR-10 without gradient descent.
PyTorch Neural Networks Bayesian Methods
Semantic Shift Analysis of PubMedBERT
OngoingYearly continual pretraining on longitudinal PubMed corpus with embedding-space, neighborhood, and concept-proximity analyses to quantify evolving biomedical knowledge.
NLP BERT Biomedical
🎓 Academic Projects
Agentic AI for Information Collection
2025Chain-of-thought LLM agent leveraging OpenAI's function-calling to orchestrate Wikipedia API tools.
LLM Agents OpenAI
Adversarial Attacks on Vision Models
2025White-box (PGD) and black-box attacks on ResNet-18 and YOLOv5 to quantify vulnerabilities.
Computer Vision Security
BioBERT Semantic Shift Detection
ResearchFine-tuning BioBERT to detect semantic shift in medical text from PubMed (1975-2025).
NLP BERT Medical AI
Deep Learning Sentiment Analysis
PublishedCOVID-19 news video sentiment analysis using LSTM, Bi-LSTM, CNN, GRU. Published at ICITA-2021.
Deep Learning NLP
Get in Touch
Let's Work Together
Interested in AI/ML research collaboration, consulting, or just want to chat about machine learning? I'd love to hear from you.
Built with Astro, TailwindCSS, and DaisyUI.
•
© 2026 Milan Varghese. All rights reserved.