Portfolio
Aaron R
Co-Founder at ZapServe • MSc Data Science & Geoinformatics Student • AI Engineer • GeoAI Researcher • Building Agentic AI and Geospatial Intelligence Systems.
Get in touch →
Portfolio
Co-Founder at ZapServe • MSc Data Science & Geoinformatics Student • AI Engineer • GeoAI Researcher • Building Agentic AI and Geospatial Intelligence Systems.
Get in touch →
I'm a Co-Founder & Director of ZapServe Interconnecting Pvt Ltd — a DPIIT Recognized startup building an AI-powered hyperlocal service platform. I combine intelligent matching, real-time location detection, and multi-agent workflows to connect users with trusted service providers.
Currently pursuing my MSc in Data Science & Geoinformatics at Digital University Kerala / IIITM-K, I work at the intersection of Agentic AI, GeoAI, and Geospatial Intelligence — building systems like GeoSI, an AI-powered conversational geospatial operating system.
My work spans LLMs, RAG, Heterogeneous Graph Neural Networks (HGNNs), computer vision for satellite imagery, and autonomous GIS workflow automation — translating cutting-edge AI research into real-world systems.
Specializing in applied data science, geospatial analytics, machine learning, GeoAI, and intelligent systems design. Research focus on Agentic AI, LLMs, and Geospatial Intelligence Systems.
Core studies in earth sciences, spatial analysis, GIS mapping, and geological field research methodologies — forming the foundation for my GeoAI and remote sensing work.
Foundation in agricultural sciences, environmental studies, and biological systems.
Leading product development, AI engineering, and platform architecture for a DPIIT Recognized hyperlocal service marketplace. Built and deployed the AI Matching Engine, managed MVP development, startup operations, fundraising preparation, and team building.
Applied GIS and remote sensing techniques for geological surveys. Conducted fieldwork, spatial data analysis, and research presentations — grounding expertise in geospatial data processing.
AI-powered conversational geospatial operating system integrating LLMs, RAG, GIS, GeoAI, and multi-agent workflows. Enables natural language interaction with spatial data, automated GIS operations, and intelligent geospatial decision support.
Heterogeneous Graph Neural Network model for credit card fraud detection achieving 96.6% accuracy and 0.79 ROC-AUC. Models transaction networks as heterogeneous graphs to identify complex fraud patterns invisible to traditional methods.
Agentic RAG system for intelligent exam preparation using LangGraph and LLMs. Autonomously retrieves relevant study material, generates contextual questions, evaluates answers, and adapts difficulty — acting as an intelligent AI study companion.
YOLO + OpenCV + PyTorch based system for automated extraction of geospatial features from satellite imagery. Detects and classifies terrain features, land use patterns, and infrastructure elements from high-resolution satellite data.
AI-based worker recommendation and hyperlocal service allocation system powering the ZapServe platform. Combines real-time location intelligence, skill matching algorithms, and demand prediction to optimally connect users with service providers.
Applied remote sensing and GIS techniques for geological field surveys — spatial data analysis, terrain modeling, raster processing, and research presentations at University of Kerala. Foundation for advanced GeoAI research.
DPIIT Recognized Startup Founder
ZapServe Interconnecting Pvt Ltd recognized under the Startup India initiative by DPIIT, Government of India.
Co-Founder & Director — ZapServe Interconnecting Pvt Ltd
Incorporated technology company building AI-powered hyperlocal services. Leading product strategy, AI engineering, and operations.
Built GeoSI — AI Powered Geospatial Operating System
Developed a conversational AI system integrating LLMs, RAG, GeoAI, and multi-agent workflows for geospatial intelligence.
Developed HGNN Fraud Detection System (96.6% Accuracy)
Designed and trained a Heterogeneous Graph Neural Network achieving 96.6% accuracy and 0.79 ROC-AUC on credit card fraud detection.
Built Agentic RAG Systems and Multi-Agent Workflows
Developed production-ready Agentic AI systems using LangGraph, CrewAI, and LangChain for intelligent automation pipelines.
Expertise in LLMs, GeoAI & Intelligent GIS Automation
Deep practical expertise spanning Large Language Models, Geospatial AI, and autonomous GIS workflow automation for earth observation and decision support.
MVP Platform Deployed
Successfully built and deployed the ZapServe minimum viable product — a full-stack AI-powered hyperlocal service marketplace.
Startup Pitching Programs
Participated in competitive startup pitching events, presenting ZapServe's vision to investors, accelerators, and mentors.
"Building intelligent systems at the intersection of Agentic AI, GeoAI, and real-world impact — where geospatial intelligence meets autonomous decision-making."