Freelancing and consultancy for maps and data. Contact Harry at projects@HKuril.com.
Make your project browsable, navigable, and exciting! Visualise relationships, notice patterns, or share information. Make use of the wealth of public information on OpenStreetMap and other sources.
Explore your datasets, fit models, and extract meaning from complexity. Display your information so it's clear and attention-grabbing. Use modern machine learning where appropriate, or classical statistical methods.
Capture the spatial aspect of your project in beautiful, clear, information-rich maps. Include the latest remote-sensing datasets including satellite imagery and digital elevation models. Inform your strategy.
Add to your existing data architecture, create a new workflow, or build a tool that your team can use, or which your customers can interact with.
I built the WildMAPS Species Distribution Model dashboard. This owner-updatable catalog of raster data is a research tool for ecologists worldwide. It’s built to display big files (100s of gigabytes) smoothly, with a low hosting cost (pennies per week).
I processed more than five million rows of mental-health reporting, for a group of NHS (UK National Health Service) psychiatrists trying to understand how different restraint methods are used. The results and visualisations revealed severe gaps in reporting, and were published in BJPsych Bulletin.
(For confidentiality reasons, no specific details are given.) In a soil-gas survey, I planned survey locations, collected field data, coordinated teams, and plotted results, to determine the prospectivity of a target region.
The SalGo Team were locked in to an expensive subscription for the map on the homepage of their plant and animal database (COMPADRE/COMADRE). In a day's work, I swapped it out for a permanent, free, improved solution, integrating with their existing PHP backend.
I'm a deployable geospatial volunteer for MapAction, with over seven weeks of emergency response experience, including field missions in Myanmar and Mozambique.
I've done than 21 weeks of geological and geophysical fieldwork in Europe, the Americas, and India, including: six weeks of solo mapping in Jamaica, and leading a gravity survey in Saint Lucia.
See my full CV for other skills: programming and GIS tools, spoken languages, teaching, writing, and even long-distance hiking, if you need trail maps...
I take pride in my products, and I'm just as invested as you in the quality of the result.
Don't get locked in to expensive contracts! Free software is now powerful enough for any geospatial task. No recurring bills, and no software overheads from me.
We'll agree a timeframe, and I can get things done fast!
I'm a good communicator and I'll iterate the product with you to make sure you get what you want. So... get in touch! Email me, Harry, at: projects@HKuril.com
I built the WildMAPS Species Distribution Model dashboard. This catalog of raster data is a research tool for ecologists worldwide. It's built to display big files (100s of gigabytes) smoothly and with very low hosting cost (pennies per week). Behind the scenes, the researchers can use the workflow I built to keep the website updated.
Tools: rasterio, shapely, geopandas, GDAL, tippecanoe, AWS S3, Maplibre-GL JS, React, OpenStreetMap
I processed more than five million rows of mental-health reporting, for a group of NHS (UK National Health Service) psychiatrists trying to understand how different restraint methods are used. The results and black-and-white visualisations revealed severe gaps in reporting, and were published in BJPsych Bulletin.
Tools: Matplotlib (PyPlot), NumPy, xarray, Excel
(For confidentiality reasons, no specific details are given.) In a soil-gas survey, I planned survey locations, collected field data, coordinated teams, and plotted results, to determine the prospectivity of a target region.
Tools: QGIS, Google Maps, Matplotlib (PyPlot), NumPy, Excel
The SalGo Team were locked in to an expensive subscription for the map on the homepage of their plant and animal database (COMPADRE/COMADRE). In a day's work, I swapped it out for a permanent, free, improved solution, integrating with their existing PHP backend.
Tools: Maplibre-GL JS, React, PHP, OpenStreetMap
Since joining UK-based charity MapAction as a geospatial volunteer, I've spent over seven weeks working emergency humanitarian responses. This has included field missions to the 2025 Myanmar earthquake and 2026 floods in southern Mozambique. Under extreme time pressure, we produce maps and manage information for all stages of the response: Search and Rescue, Rapid Needs Assessment, and Distribution. We make use of all sorts of public datasets: OpenStreetMap, gridded population, building extents, and more.
Tools: ArcGIS Pro, QGIS, Python, OpenStreetMap
(For confidentiality, no specific details are given.) As a data scientist at startup OKKO Health, I created an algorithm to detect vision loss, based on user data. The algorithm integrated with the existing data pipeline for daily updates.
Tools: Python, Numpy, Pandas, Matplotlib, SQL
(For confidentiality, no specific details are given.) A conservation organisation, active in the Amazon rainforest, requested cartography (in Spanish) at various scales: a map of their headquarters, a tree survey, cadastral maps, and maps of spiritual sites.
Tools: QGIS, Inkscape
As a passion project, I made maps to illustrate the best locations for viewing the aurora borealis. These were #3 in Maps.com's Most Popular Maps of 2025.
Tools: rasterio, QGIS, scipy
As a personal project, I built an interactive map of the Transcaucasian Trail (TCT), a long-distance walking route which crosses Armenia and Georgia, as well as a live Garmin satellite tracker map. I’m on the TCT Association Hikers’ Board where I help with GIS, trail notes, and other aspects of trail development.
Tools: Leaflet, Folium, Protomaps (PMTiles)
During my PhD, I used seismic tomography to study the Earth’s deep interior, and tested methods of modelling Earth’s lowest-frequency vibrations. Read more in my PhD thesis.
Tools: Seismology, Geophysics, Modelling, Signal processing, High-performance computing, Python
I produced a map to showcase Alice Morrison’s 2,200-km (1,370-mile) foot journey across Saudi Arabia.
Tools: QGIS, Inkscape
I developed a set of Python codes to process GPX files, and tested them on the recordings of Russ Cook's remarkable run, from south to north across Africa. See the nerdy post on my personal blog.
Tools: gpxpy, web scraping, selenium, beautifulsoup, Inkscape, QGIS, MatPlotLib