Every project starts from an idea and raw data β satellite imagery, census records, transit feeds, sensor streams. I wrangle it, model it, and ship it as platforms, CLIs, and APIs that real users depend on. Three live products. Open-source tools on PyPI. Automation saving hundreds of hours annually for research and government across Australia.
From raw data to production platforms β measurable outcomes for research, government, and enterprise
Investigated LLM spatial reasoning across 535 questions. Found 0% accuracy on distance calculations, ~80% failure on topology. Designed the GeoSpark Protocol giving AI models ground-truth spatial computation. Published on PyPI with 446 tests. Open source, Apache 2.0.
GeoSpark Β· Open Source Β· pip install geospark-aiBuilt a platform matching satellite, LiDAR, and aerial data to project needs from plain English. 80% handled by PostGIS + structured lookups; LLM only when confidence < 0.7. Shipped as web app, pip-installable CLI (20 commands), MCP server (6 tools), and REST API.
Terrascout Β· terrascout.app Β· 217 testsIntegrated ABS Census (14 dataflows), VicMap, PTV GTFS (30K+ stops), Microsoft Building Footprints (5M+), and 20+ more sources into 23 analysis models calibrated to Australian planning standards. QGIS plugin for professionals, web platform for teams.
UrbanMind Β· urbanmind.terrascout.appNon-technical researchers couldn't query PostGIS databases without the GIS team. Built an AI platform that translates plain English into spatial queries β reducing support requests by 70% with 90% accuracy. In production at University of Melbourne.
UMAMI Β· University of Melbourne50,000+ bi-weekly property records validated against 339 database tables. Automated the entire M1 form processing pipeline β from data extraction through validation to submission. Eliminated manual errors completely.
Greater Shepparton Council Β· VictoriaDeployed mobile GIS across flood and fire-affected regions with offline sync, standardized damage assessment forms, and real-time dashboards for leadership. Cut reporting lag from days to hours.
Parks Victoria Β· Emergency ResponseOrganizations I've Worked With
End-to-end platforms β from research question to production deployment
Enterprise GeoAI platform with 6-agent swarm intelligence, 11+ MCP servers, visual workflow editor, IoT dashboard, and RAG-powered semantic search.
Provider-agnostic satellite, LiDAR, and aerial data scoping platform. Describe your project in plain English, get ranked sensor recommendations with feasibility scores. 217 tests, 5 AI agents, 6 MCP tools.
Comprehensive urban planning platform with 23 analysis models, 36+ real data sources, and AI-powered insights β calibrated for Australian practice. QGIS Plugin + Web Platform with 7 completed phases.
Open-source protocol giving AI models ground-truth spatial reasoning. 535 benchmark questions prove LLMs fail 70%+ on spatial tasks β GeoSpark fixes this with real topology and geodesic computation.
WebODM plugin automating KSA-GRF17 coordinate system configuration for drone imagery in Saudi Arabia. Auto-detects UTM zone from EXIF GPS data and injects the correct PROJ string β zero configuration.
GEE analysis for Venice Lagoon habitat mapping using Sentinel-2 imagery, vegetation indices (NDVI, SAVI, NDMI, NDPI), and supervised ML classification for species restoration.
Real-time monitoring system for indoor environments using Digital Twin technology and IoT sensors. Four-layer architecture tracking COβ, temperature, and humidity with LoRaWAN connectivity.
Automated ETL pipeline processing 50,000+ property records bi-weekly, reducing processing time by 90% and eliminating manual errors.
Full-stack web application visualizing environmental monitoring data across 200+ sensor locations with real-time updates.
Mobile GIS solution enabling rapid damage assessment for flood and fire recovery operations with offline capability.
Natural language interface for complex geospatial database queries, reducing support requests by 70%.
Scalable cloud infrastructure for asset management with automated data synchronization and secure API endpoints.
AI-powered disaster response platform with 3 autonomous agents for real-time disaster detection, claims verification, and automated client outreach.
Interactive dashboard tracking Sustainability Plan 2030 targets with 10+ widget types, visual editor, and role-based access control.
Full-stack interface for managing 30+ Model Context Protocol servers with visual workflow builder, GeoAI analysis, and Gemini-powered query planning.
Continuous learning and building innovative geospatial solutions
Full-stack capabilities β from raw spatial data through AI integration to production deployment
Spatial Data Engineer & GeoAI Specialist
I'm a Spatial Data Engineer and GeoAI Specialist with a Master's in Spatial Engineering from the University of Melbourne. I build AI-powered geospatial platforms end-to-end β from wrangling raw data (census records, satellite imagery, transit feeds, sensor streams) through spatial modelling and AI integration to production deployment with CLIs, APIs, and MCP tools.
Currently GIS Officer at Greater Shepparton Council and Data Engineer at CSDILA (University of Melbourne), where I've shipped three live platforms (Terrascout, UrbanMind, UMAMI), published an open-source spatial AI library on PyPI (GeoSpark), and automated workflows saving hundreds of hours annually for government and research teams.
I'm driven by a simple pattern: find a real problem in the geospatial domain, figure out what data it needs, build the intelligence layer, and ship it as a tool anyone can use.
University of Melbourne
Feb 2022 - Dec 2024
Capstone: Digital Twin with IoT for real-time indoor & outdoor environmental monitoring
Azad University Central Tehran
Sep 2017 - Jul 2020
Oil and Gas Exploration specialization
Azad University Bandar-E-Anzali
Sep 2013 - May 2017
Greater Shepparton Council
Centre for Spatial Data Infrastructures (CSDILA)
Parks Victoria
5+ roles across government, research, and private sectors showcasing versatility and impact
Greater Shepparton Council
Leading enterprise GIS operations for local government, specializing in cadastral data automation and property information systems.
Centre for Spatial Data Infrastructures and Land Administration
Architecting cutting-edge GeoAI platforms and spatial data engineering solutions for University of Melbourne research initiatives.
Mansfield Shire Council
Delivered field data collection solutions and staff training for local government operations.
Centre for Spatial Data Infrastructures and Land Administration
Contributed to spatial data infrastructure research, supporting the development of digital twin technologies and spatial data management frameworks.
Curtin University
Provided specialized GIS analysis and data support for university research projects.
Parks Victoria
Led critical GIS operations for Victoria's largest land manager, specializing in emergency response mapping and environmental risk assessment.
HUMAYA Automation
Developed enterprise web mapping applications for infrastructure management and spatial analysis.
Avale Enterprises
Specialized in satellite imagery analysis and machine learning for environmental monitoring and land use classification.
Professional Roles
Flood & Fire Disaster Mapping
Roles at Peak
Government, Research, Private
79+ verified credentials across AI, GIS platforms, cloud computing, and data engineering
Showing 74 of 74 professional certifications
Have a spatial data challenge? Let's explore how I might be able to help.