Ajaypal Singh

• Project Leadership & Team Management - Served as team lead and project coordinator across multiple geospatial agriculture initiatives, managing cross-functional workflows, coordinating team activities, assigning technical responsibilities, leading project discussions, and ensuring timely execution of deliverables. Worked closely with stakeholders to align business objectives with agricultural technology requirements while maintaining smooth communication between development, research, and operational teams.

• Crop Health Monitoring & Automated Satellite Data Processing Pipeline - Designed and implemented a unified end-to-end geospatial agriculture framework for real-time crop monitoring and satellite image processing. Developed automated pipelines for satellite data download, ingestion, band extraction, and multi-spectral analysis using Python-based custom formulas. Generated crop health mosaics and analytical outputs to support agricultural decision-making and improve field-level monitoring. Integrated APIs for visualization and result delivery, while containerizing the solution with Docker for scalable deployment on servers. Built the system to handle large-scale geospatial datasets efficiently, ensuring production readiness, reliability, and automation-driven processing workflows.

• Satellite-Based Crop Type Classification & Prediction Project - Led development of an agricultural crop classification model using satellite imagery combined with field-level data. Extracted geospatial and image-based features, performed image preprocessing, trained machine learning and CNN-based models for crop identification, and achieved over 60% classification accuracy. Produced raster and vector outputs (.shp files), configured Geoserver for data publishing and visualization, and built APIs for displaying results in an interactive environment.

• AI Chatbot for Geospatial Agricultural Queries - Designed and deployed intelligent chatbot systems focused on geospatial agriculture use cases. Integrated vector databases using Postgres for semantic search and retrieval, implemented prompt engineering strategies for improved interaction quality, and enabled natural language querying of crop, field, and geospatial datasets. Improved accessibility of agricultural information for users through conversational AI workflows.

• Computer Vision for Live Stream & Image Processing - Developed OpenCV-based computer vision workflows for agriculture-related image and live stream analysis. Applied image preprocessing, object detection, and frame analysis techniques to support real-time monitoring tasks in field environments. Worked on dynamic image-processing pipelines to improve automation and visual interpretation of agricultural scenarios.

• Machine Learning Model Training & Deployment for Agriculture - Built and deployed predictive models tailored to agricultural use cases, including forecasting models, CNN-based image analysis, and YOLO-based object detection solutions. Implemented model training, evaluation, and deployment workflows while following basic MLOps practices for versioning, testing, and monitoring. Focused on improving prediction accuracy and enabling scalable deployment in production environments.

• Tool Ecosystem & Geospatial Analysis Support - Utilized a broad ecosystem of tools to support experimentation, validation, and deployment across multiple projects. Worked extensively with Google Colab for prototyping, Postman for API testing, QGIS for mapping and spatial analysis, SQL for querying datasets, and geospatial utilities for raster/vector data processing. Contributed to improving productivity, validation accuracy, and overall workflow efficiency.