"In God we trust, all others must bring data." – W. Edwards Deming
class DataScientist:
def __init__(self):
self.name = "Shahid Ul Islam"
self.role = "Data Scientist | ML Engineer"
self.location = "Kulgam, Kashmir"
self.education = "MCA — Islamic University of Science & Technology (CGPA: 9.24)"
self.languages = ["en_US", "ur_PK"]
self.email = "shahid9664@gmail.com"
def say_hi(self):
print("Thanks for visiting! Let's build data-driven solutions together.")
def current_focus(self):
return [
"🏥 Healthcare AI & Clinical Decision Support",
"🤖 Agentic & Multi-Agent AI Systems",
"🧠 Deep Learning — CNNs & Vision Transformers",
"📊 Predictive Modeling & Feature Engineering",
"🔤 Natural Language Processing & Text Classification",
]
def expertise(self):
return {
"ml_ai" : ["Supervised/Unsupervised Learning", "Transfer Learning",
"Class Imbalance Handling", "Model Evaluation"],
"deep_learn" : ["CNNs", "Vision Transformers (ViT)", "NLP", "TF-IDF",
"Sentiment Analysis"],
"agentic" : ["Multi-Agent Architectures", "Tool-Augmented AI",
"Deterministic Reasoning", "Safety-Critical AI Design"],
"cloud" : ["Google Cloud Platform", "BigQuery", "Looker"],
}
me = DataScientist()
me.say_hi()
|
Multi-class chest X-ray classifier detecting COVID-19, viral & bacterial pneumonia using transfer learning. Applied VGG16 and Vision Transformers (ViT-B/16) on 6,500+ medical images, achieving ~86% accuracy under class imbalance. Evaluated with precision, recall, F1-score, and confusion matrices. And performed Quantitative Faithfulness comparison between CNN's and ViT's using the XAI. Stack: Python · PyTorch · TensorFlow · CNNs · ViT · Augmentation |
End-to-end ML pipeline classifying tumors as benign or malignant with ~98% accuracy. Performed feature engineering and data preprocessing, improving baseline performance by ~15%. Deployed as an interactive Streamlit web application for real-time clinical predictions. Stack: Python · Scikit-Learn · Random Forest · Streamlit |
|
Enterprise-grade analytics platform analyzing Apple hardware vs. software revenue trends. Built an NLP-driven Executive Copilot and live telemetry simulator using Random Forest regression and Autoregressive pipelines. Integrated unsupervised clustering (K-Means, DBSCAN) and SciPy optimization (SLSQP) to translate data anomalies into quantified strategic directives. Stack: Python · Pandas · Plotly · Streamlit · Scikit-Learn · NLP |
|
| Certificate | Issuer |
|---|---|
| Machine Learning | Columbia University |
| Advanced Learning Algorithms | Coursera (DeepLearning.AI) |
| Supervised ML: Regression & Classification | Coursera (DeepLearning.AI) |
| Unsupervised Learning, Recommenders & Reinforcement Learning | Coursera (DeepLearning.AI) |
| Introduction to Generative AI | |
| Introduction to Large Language Models | |
| Analyzing & Visualizing Data in Looker |
From Khanz9664 | Thanks for visiting!