About me
I’m Samanwoy Saha, a software engineer now focusing on machine learning. I’m building end-to-end solutions that integrate machine learning models with robust software systems.
Skills
- Languages -
C/C++JavscriptTypescriptPythonSQL - Frontend -
HTMLCSStailwindReactNext.jsStreamlit - Backend -
Node.jsExpress.jsPrisma ORMWebSocket - Database -
PostgreSQLMongoDB - Cloud & Devops -
AWS (EC2, S3)LinuxDockerCI/CD Pipelines - Data Science & Machine Learning:
Feature EngineeringStatistical AnalysisExploratory Data Analysis (EDA)Data PreprocessingModel EvaluationModel DeploymentLinear RegressionLogistic RegressionSupport Vector MachineK-Means ClusteringScikit-LearnPandasNumPyMatplotlibSeaborn - Tools -
GitGithubJupyter Notebook - CS Subjects -
DBMSComputer NetworkOperating SystemObeject Oriented ProgrammingData Structures and Algorithms
Recent Projects
🏥 Multiple Disease Prediction
- Website | Github
- Tech Stack: Python, Streamlit, Pandas, Numpy, Matplotlib, Seaborn, Scikit Learn, Github Actions, Docker, AWS EC2
- Tech Stack: Python, Streamlit, Pandas, Numpy, Scikit Learn, Github Actions, Docker, AWS EC2
- Algorithms Used: Diabetes(Support Vector Machine), Heart Disease(Logistic Regression), Parkinson’s Disease(Support Vector Machine)
- Model Accuracy: Diabetes(77%), Heart Disease(78%), Parkinson’s Disease(87%)
- Automated CI/CD Pipeline: GitHub Actions workflow with testing, Docker builds, and AWS EC2 deployment
Customer Segmentation
- Website | Github
- Tech Stack: Python, Streamlit, Pandas, Numpy, Matplotlib, Seaborn, Scikit Learn
- Algorithms Used: K-Means Clustering (Elbow Method)
- Statistical summary of customer data, Distribution analysis of key features, Correlation analysis between variables, Visualization of customer patterns are done through EDA
- Purchasing patterns and customer preferences which can provide actionable insights for marketing and sales teams and enable targeted campaigns and personalized customer experiences
Contact
Send me a mail to samanwoy.saha1@gmail.com