About
Software Engineer & Data Analyst with experience building data-intensive platforms, distributed data pipelines, and analytics systems. Skilled in Python, PySpark, Databricks, SQL, Django, and React, with experience processing large-scale structured and unstructured datasets. Currently developing a data-driven crop risk analytics system leveraging satellite imagery, radar signals, and weather datasets to engineer predictive features for crop damage detection and insurance risk modeling. Experienced in feature engineering, data pipelines, and scalable analytics workflows.
Skills & Expertise (30)
Work Experience
Analyst
Virtusa
Sep 2024 - Aug 2025
Analyzed large-scale advertising datasets to optimize Ads campaign performance. Used Excel (advanced formulas, pivot tables) and Python (Pandas) for data processing and analysis. Built Power BI dashboards to visualize KPIs such as Handling Time and First Pass Yield (FPY). Conducted daily trend and anomaly analysis, delivering actionable insights. Prepared detailed reports and presentations for stakeholders, improving decision-making efficiency.
Data Analyst
Dvara E-Registry
Aug 2025 - Present
Built a PySpark-based data processing pipeline on Databricks to analyze large-scale multi-source datasets including satellite imagery and agricultural signals. Developed workflows to extract spectral and radar signal values from satellite datasets and transform them into structured analytical features. Implemented pixel-level aggregation and heatmap generation pipelines to monitor crop stress signals across farmer land parcels. Engineered early-stage damage detection logic using multi-band signal combinations to identify crop anomalies and potential yield loss. Designed scalable pipelines to map satellite signals to farmer-level datasets using WKT farm boundaries and centroid-based extraction methods. Currently integrating weather datasets (rainfall, temperature, humidity) to build multi-variable feature sets for crop risk modeling. Performing feature engineering across temporal satellite signals and weather parameters to support predictive modeling for crop damage detection. The system is currently in Proof-of-Concept (POC) stage, aimed at building a data-driven crop insurance risk assessment engine. Developed an end-to-end Digitized Plot Data Report featuring multi-level drill-downs across Plot, Village, District, and State levels to monitor regional agricultural health and Engineered complex logic within reports to aggregate CreditScores, identifying and summing eligible plots for loan credit while visualizing Crop History and farm risk factors. Integrated and visualized datasets regarding proximity to markets and water bodies to provide a 360-degree view of farm productivity and environmental risk. Architected the KhetScore reporting engine to facilitate agricultural credit assessments for over 10,000 farmers. Developed automated Python-based data pipelines to transform raw PostgreSQL data into structured JSON, enhancing reporting accuracy. Optimized data processing workflows using Pandas and SQL, resulting in a 35-40% increase in computational efficiency. Integrated AI-assisted development tools (GitHub Copilot, Claude) to accelerate feature delivery cycles by 30%. Engineered a geospatial validation framework for MOSDAC, validating 100,000+ satellite records for consistency and accuracy. Automated QA protocols for satellite data, reducing manual validation time by 60% and improving downstream model accuracy by 25%. Redesigned and modernized the company website and built a full-stack HRMS platform to digitize internal HR and operational processes. Developed a role-based HRMS application using React (frontend), Django (backend), PostgreSQL (database) with RESTful APIs. Implemented hierarchical access control with distinct dashboards and permissions for Employee level, Manager level, and HR level. Built interactive dashboards displaying consolidated employee data including personal details, policies, certifications, leave balances, and benefits. Implemented secure authentication and authorization mechanisms to ensure data privacy and role-based access. Enabled 100% digital employee onboarding, reducing manual HR workload by ~60%. Improved onboarding and request turnaround time by ~45–50%. Optimized backend APIs and database queries, improving application response time by ~30%. Used AI-assisted coding tools (GitHub Copilot, Claude, Cursor) to accelerate development, refactoring, and API design.
Education
Bachelor of Science (Agriculture) - Lovely Professional University
- 2024 · Afghanistan