I am a motivated and detail-oriented Full Stack Developer with strong expertise in HTML5, CSS3, JavaScript, React, Node.js, Express, and MongoDB (MERN stack). I build responsive, user-friendly web applications and tailor solutions to project requirements.
I also have a solid background in data science, predictive modelling, and visualization, enabling me to deliver data-driven, scalable applications. I work well independently in remote environments and collaborate effectively to meet tight deadlines.
Skills
Experience Level
Language
Work Experience
Education
Qualifications
Industry Experience
• Employed Python with SQLAlchemy to extract and analyse data from AdventureWorks2022.
• Generated descriptive summaries and visualizations of key database attributes using Pandas and
Matplotlib.
• Conducted statistical analysis, including confidence intervals, to derive meaningful insights.
• Provided actionable recommendations based on vendor performance and pricing data.
• Developed a Streamlit app for digit recognition using ML models trained on the MNIST dataset.
• Achieved highest accuracy with SVM model at 97.64%, followed by Random Forest and KNN.
• Implemented custom image preprocessing and model fine-tuning to improve prediction accuracy.
• Enabled user-friendly data collection through image capture with web cameras or mobile devices.
• Developed linear regression models in R to forecast Volvo V60 prices.
• Incorporated predictor variables including Model Year, Mileage, Fuel Type, Gearbox, and
Horsepower.
• The model accounted for 84.3% of the price variance, with respective RMSE values of 32,170.44
(validation) and 41,046.34 (test).
• Recognized the necessity for additional refinement owing to the Gearbox’s marginal significance.
• Developed CNN models for face expression, age, and gender recognition.
• Utilized Python, TensorFlow, and Keras for data preprocessing, model training, and evaluation.
• Achieved accuracies of 77% (Expression), 95% (Gender), and MAE of 3.5080 (Age) with CNN
models.
• Developed a modular Python ETL pipeline to extract, transform, and load Big Mac index data
from CSV to SQLite.
• Utilized pandas for data cleaning and sqlite3 for database integration.
• Implemented structured logging and exception handling for robust execution.
• Built automated testing with unittest and scheduled execution via Windows Task Scheduler.
• Delivered a maintainable, well-documented codebase with clear module separation.
• Built a robust forecasting pipeline using ensemble models (Linear, Random Forest, Gradient
Boosting, Voting Regressor) and LSTM to predict birth rates and education costs.
• Integrated demographic and cost data from SCB into a structured SQLite database using pandas
and sqlite3.
• Achieved high model performance (R² > 0.95) with rigorous validation (RMSE, MAE, cross
validation).
• Deployed insights via an interactive Streamlit dashboard to support data-driven education
budgeting.
Thesis – Predictive Modeling & App Development for Mental Health
TrichMind: Relapse Risk Prediction for Trichotillomania
• Developed a machine learning pipeline (Logistic Regression, Random Forest, XGBoost) to predict
relapsed risk based on emotional and behavioral user-reported data.
• Collected and preprocessed survey and forum data; identified key triggers like stress, anxiety,
and solitude.
• Built and deployed a Streamlit-based web/mobile prototype to deliver real-time risk feedback and
personalized coping strategies.
• Developed a full-stack social media platform using MERN stack.
• Implemented authentication, CRUD operations, and real-time features.
• Applied best practices in modular design, API development, and responsive
UI.
Designed and developed a travel cost calculator with UI/UX principles and Agile
methodology.
Developed a JavaScript-based shopping cart feature with dynamic DOM
manipulation.
• Developed interactive dashboards using Adventure Works Cycles, a relational database.
• Created a star or snowflake schema data model integrating vendor performance metrics: Total
Orders, Total Revenue, Average Lead Time, Quality Score, and On-Time Delivery Rate.
• Visualized sales trends over time and identified top and bottom-performing products.
• Implemented advanced features like drill-down hierarchies, drill-through, conditional formatting,
and custom tooltips.
• Created a user-friendly, multi-page report with a custom theme.
Hire a Data Scientist
We have the best data scientist experts on Twine. Hire a data scientist in Gothenburg today.