Dublin → Wrocław · Available for new work

Software for the real world, mapped.

I'm Dean Sherwin, a Python and Django engineer working where data meets geography — healthcare systems, rail networks, and the fields that feed us.

A retro computer displaying a globe, surrounded by maps and survey notes
From the survey to the satellite.

What I do

Three tools, one approach

01

Backend engineering

Django and Python services built to be understood six months later: explicit boundaries, sensible data models, and tests around the parts that cost money when they break.

02

Geospatial systems

PostGIS data models, tiling pipelines and map interfaces that turn survey points, satellite scenes and field records into something a human can act on.

03

Product delivery

Small teams, short feedback loops, and a working demo over a polished slide deck. I have led remote teams across time zones and shipped through the messy middle.

Track record

Seven years, three domains

Software work
Healthcare

Clinical systems that clinicians trust

Full-stack work on patient-facing and internal medical platforms, with an emphasis on auditability, access control and interfaces that stay out of the way of care.

Rail & surveying

Web-GIS for engineering projects

Project management backends and client-facing maps for railway surveying work, including delivery for Irish Rail and a long list of very large drawings.

AgTech

Earth observation for growers

Crop modelling, yield prediction and decision-support tools that translate satellite data into recommendations a farmer can use before the season runs out.

Teams

Remote by default, senior by habit

Leading distributed teams, reviewing code kindly but carefully, and keeping delivery predictable without turning the process into a second job.

Principles

How I work

  1. Start with the constraint Find the real bottleneck before choosing the tool.

    Bandwidth, a legacy schema, a regulatory deadline — the constraint decides the architecture, not the other way around.

  2. Make it observable If you cannot see it failing, you do not own it.

    Logs, metrics and honest error states are part of the feature, not an afterthought for the week before launch.

  3. Write it down Decisions deserve a paragraph.

    Short design notes and useful commit messages save more time than clever abstractions ever will.

  4. Ship, then learn A small release beats a perfect plan.

    Get it in front of the people who will use it, watch what happens, and let the next iteration be informed by reality.

Writing

Notes from the workbench

All posts

Next project

Have a problem worth solving?

I'm open to backend, geospatial and data-heavy product work — remote or around Wrocław. Tell me what you're building.

Get in touch