I am an Urban Data Scientist and GIS Analyst with a strong foundation in spatial data science, advanced analytics, and urban systems modeling. My expertise lies in combining geospatial technologies, data science, and domain knowledge to solve complex challenges in transportation, urban mobility, and sustainability.
With a Master's in Urban Data Science and Analytics from the University of Leeds and a background in Geoinformatics, I bring a rigorous analytical mindset to urban and environmental problems. Over the past years, I have worked on projects spanning consumer behavior segmentation, urban carbon storage estimation, accessibility analysis, and predictive congestion modeling.
My approach integrates traditional spatial analysis with modern machine learning and cloud-based automation, allowing for data-driven solutions that are not only technically sound but also practically impactful.
- Urban mobility and transport optimization
- Traffic congestion prediction and mitigation
- Geospatial modeling for sustainability and climate resilience
- Integrating real-time data streams (IoT, APIs) into decision-making
- Automation of ETL workflows using Airflow, GitHub Actions, and cloud tools
- Programming: Python (Pandas, NumPy, Scikit-learn, Plotly, Dash), R, SQL
- GIS & Remote Sensing: ArcGIS, QGIS, PostGIS, Google Earth Engine
- Data Engineering & Automation: Apache Airflow, Docker, GitHub Actions, Bash, ETL pipelines
- Visualization & BI: Tableau, Power BI, Matplotlib, Seaborn
- Cloud & CI/CD: Git, GitHub, GitHub Actions (CI/CD), in progress: GCP
-
Urban Congestion Prediction & Optimization (King William Street, London)
- Developed a live data pipeline integrating bus arrival times, line status, disruptions, and weather data to monitor and predict congestion patterns.
- Automated collection and version-controlled storage via GitHub Actions to enable real-time, continuous updates and future model training.
-
UK Road Accident Severity Forecasting & Risk Factor Analysis
- Analysed 1.7M+ road accident records (2005–2015) from the UK Department for Transport, integrating spatial (collision coordinates, road type) and non-spatial (driver age, vehicle type, weather, lighting) features to understand key determinants of accident severity.
- Developed SARIMAX-based time series models to forecast future accident trends, validating against 2016–2019 data; findings revealed declining trends attributed to policy changes and public safety campaigns.
-
Consumer Segmentation for Northern Railway Stations
- Mapped and analysed footfall data for ~3.6 million visitors across Harrogate, St. Helen Central, and Worksop stations, using geofencing and 12-month mobility data to delineate primary, secondary, and tertiary catchment zones.
- Implemented advanced spectral and k-means clustering on Mosaic demographic and behavioural segments, uncovering critical consumer profiles (e.g., “Affluent Bargain Hunters,” “Budget-Conscious Commuters”), which informed station-specific retail strategies.
I am passionate about transforming urban data into actionable insights that drive better, more sustainable decisions. I enjoy collaborating on projects at the intersection of cities, data science, and technology — whether it is developing advanced spatial models, building automated data pipelines, or visualizing complex urban dynamics.
I am actively seeking opportunities to contribute to innovative teams working on smart cities, urban mobility, and geospatial analytics, where I can continue to learn and make a meaningful impact.