classRajdeepDas:
def__init__(self):
self.role="Data Science / Machine Learning Student"self.focus= ["Predictive Modeling", "Survival Analysis", "EDA"]
self.stack= ["Python", "pandas", "scikit-learn", "SQL"]
self.learning= ["Deep Learning", "MLOps", "Feature Engineering"]
defcurrent_goal(self):
return"Build end-to-end ML projects that solve real problems"- 🔬 I build machine learning projects that turn messy, real-world datasets into models people can act on.
- 📊 Comfortable across the full pipeline — EDA → preprocessing → modeling → evaluation → interpretation.
- 🧠 Currently deepening my work in healthcare analytics and classification modeling.
- 📫 Reach me at dasrajdeep346@gmail.com
Languages & Core
Data Science & Machine Learning
Visualization & Environment
Predicts 10-year mortality in breast cancer patients by combining survival analysis with supervised classification.
2,509 patients · 34 features · best accuracy 99.60% ➡️ View project | Predicts whether a data scientist is likely to seek a job change, built on HR Analytics data.
19,158 records · ROC-AUC 0.807 ➡️ View project |
| Area | What that looks like in practice |
|---|---|
| 🔍 Exploratory Data Analysis | Distribution profiling, correlation heatmaps, missing-value strategy, outlier handling |
| 🧹 Data Preprocessing | Imputation, encoding, scaling, deduplication, leakage checks |
| 🤖 Predictive Modeling | Logistic Regression, SVM, Decision Trees, LightGBM, model comparison |
| 📉 Survival Analysis | Kaplan-Meier estimation, Cox Proportional Hazards, hazard ratios |
| 📊 Evaluation & Interpretation | ROC-AUC, precision/recall, confusion matrices, feature importance |
| Area | Focus |
|---|---|
| 🧠 Deep Learning | Neural networks, TensorFlow / PyTorch fundamentals |
| ⚙️ MLOps | Model deployment, versioning, reproducible pipelines |
| 🗄️ SQL & Data Engineering | Query optimization, warehousing, ETL |
| 📈 Advanced Statistics | Causal inference, experiment design, Bayesian methods |
I'm open to collaboration on data science projects, research, and internship opportunities.
"Without data, you're just another person with an opinion."
— W. Edwards Deming
