Geospatial & Data Engineer
GeoAI · Spatial Intelligence · Python · SQL · Machine Learning · Enterprise GIS · Cloud & Security
Trusted data systems. Better decisions.
I build geospatial and data systems that transform complex information into trusted intelligence for public health, emergency response, and organizational decision-making.
My work connects enterprise GIS, spatial data engineering, Python/SQL automation, ETL/ELT pipelines, machine learning, and interactive decision support—with cloud infrastructure, access controls, and data governance supporting the analytical process.
My professional journey began during the 2014 Ebola response in Rivers State, Nigeria, followed by an African Union deployment to Liberia.
Professional Website · LinkedIn · Email
| Measure | Impact |
|---|---|
| Experience | 10+ years of professional experience across GIS, data systems, public-health intelligence, and spatial decision support |
| Delivery | 6,000+ maps, dashboards, and analytical products |
| Capacity building | 400+ professionals trained in GIS, spatial analysis, and digital data collection |
| Research | 10+ peer-reviewed scientific publications |
| Public service | National-level public-health surveillance and emergency-response support in Liberia |
An implemented and evaluated research prototype connecting governed surveillance data, machine-learning risk classification, spatial intelligence, explainability, authenticated APIs, and interactive decision support.
Developed for retrospective surveillance prioritization using a modelling-ready dataset of 4,760 district-month observations across 136 Liberian districts.
Read the engineering case study →
| Evidence state | Scope |
|---|---|
| Implemented and evaluated prototype | CSV/GeoJSON data layer, data preparation, comparative risk modelling, spatial analysis, SHAP explanations, FastAPI/JWT/RBAC, audit logging, and Streamlit/Folium decision support |
| Separately validated integration | PostgreSQL/PostGIS ingestion, SQL queries, spatial geometry, and ArcGIS/Python links—not represented as the public application's active backend |
| Demonstrated deployment | Docker packaging, AWS EC2 API testing, and Streamlit Community Cloud dashboard deployment |
| Proposed production pathway | Automated ingestion, streaming/distributed processing, telemetry and drift monitoring, API TLS, managed secrets, and resilience testing |
- Logistic Regression, Random Forest, and XGBoost risk modelling
- Global Moran’s I, Local Moran’s I/LISA, and Getis-Ord Gi* spatial analysis
- Distinction between observed incidence and model-derived risk
- SHAP-based global and local model explanations
- Interactive GIS maps, surveillance indicators, and relative-risk rankings
Technology stack: Python GeoPandas PySAL Scikit-learn XGBoost SHAP FastAPI Streamlit Folium Docker AWS JWT RBAC
The case study reports the original dissertation evaluation separately from post-dissertation grouped and temporal validation.
Research scope: retrospective risk classification and surveillance prioritization. Not external or prospective validation; not calibrated absolute outbreak risk. Demonstrated deployment is a functional proof of concept, not production assurance. EC2 API testing used HTTP; API TLS remains part of the production-hardening pathway.
Explore the project:
- Enterprise GIS and spatial data-management workflows
- Geospatial data validation, QA/QC, and automation
- Data-engineering and ETL/ELT pipelines
- GeoAI, spatial machine learning, and explainable analytical applications
- Interactive dashboards and geospatial decision-support systems
- Authenticated APIs and governance-aware data services
- Reproducible research software and technical documentation
My professional website is the primary home for my experience, selected projects, credentials, and case studies. GitHub provides the supporting code and technical evidence.
Explore Selected Work · Portfolio Evidence Repository
Geospatial Engineering & Enterprise GIS
ArcGIS Pro ArcGIS Online QGIS PostGIS GeoPandas PySAL Spatial ETL GIS QA/QC Spatial Statistics Cartography Remote Sensing
Data Engineering & Analytics
Python SQL Pandas NumPy PostgreSQL ETL/ELT Data Validation Feature Engineering Spark PySpark Hive
GeoAI & Explainable Machine Learning
Scikit-learn XGBoost Random Forest Logistic Regression SHAP Model Evaluation Spatial Machine Learning
Cloud, APIs & Supporting Systems
AWS Linux Docker FastAPI REST APIs Streamlit Folium Git/GitHub JWT RBAC Audit Logging
Tools span different projects and professional workflows; each case study identifies its implemented stack and evidence boundaries.
My work spans enterprise GIS, spatial data management, public-health surveillance, emergency response, and applied research.
Selected areas of contribution include:
- GIS implementation and spatial data infrastructure
- Surveillance and analytical support for Ebola, COVID-19, malaria, measles, cholera, mpox, and Lassa fever
- Dashboard development and operational reporting
- GIS training and professional capacity building
- Geospatial data quality assurance and automation
- Spatial epidemiology and public-health analytics
My work has supported programs involving AFENET, the African Union, WHO, and CDC.
University of East London / UNICAF
Dissertation submitted for assessment:
Design and Evaluation of a Privacy-Preserving GeoAI Health Surveillance System Using a Hybrid Cloud Architecture
Using Design Science Research, the study designed, implemented, and evaluated a research prototype across predictive performance, spatial intelligence, explainability, security and governance, usability, and decision-support value.
The dissertation title describes the research framing; the linked case study documents what was implemented, evaluated, separately validated, demonstrated, and proposed.
Federal University of Technology, Akure, Nigeria / UN-ARCSSTEE
Specialized training in GIS, remote sensing, spatial analysis, geospatial data management, cartography, and environmental modelling.
I have authored and co-authored 10+ peer-reviewed publications covering infectious-disease surveillance, spatial epidemiology, GIS, GeoAI, and public-health intelligence.
Google Scholar · ORCID · ResearchGate
I welcome opportunities and collaborations where geospatial and data engineering can turn complex information into trusted, actionable intelligence.
- Website: godwineakpan.com
- LinkedIn: Godwin Etim Akpan
- Email: godwinea.ai@gmail.com
Trusted data systems. Better decisions.