Geospatial Data Science opportunities

I’m looking for guidance on how to land geospatial data science remote opportunities. What skills should I focus on? How can I get a foot into this roles? Kindly advice.

Hi @Koigi ,

I think the first part of my answer to that is a question back to you: “why do want to work in this area?” What I mean is it’s worth knowing what interests you and focusing on roles that would give you opportunities to work on that. You don’t mention ML and you don’t mention remote sensing data, so you could work on geospatial data science using pretty much mostly vector data, in which case get familiar with the geopandas API and basic geospatial concepts such as CRS and projected coordinate reference systems etc. But if you really want to work on exploiting satellite data, then getting comfortable with xarray is probably time well spent.

For me, I came from an ML background and so had to learn geospatial tooling in order to get data (ground truth, typically vector data, and EO data, typically raster) into a format suitable for ML (i.e. rows of observations containing features) and then back again (to display the output in context on a map).

In order to “get a foot into” such a role, I would definitely advocate for a strong core of fundamental data science skills, so data cleaning, EDA etc. An additional dimension involved with geospatial data is just that - geospatial. It’s amazing how often there are errors in geometries, either missing or incorrect CRS or overlapping geometries that you weren’t expecting. Even to get started, being able to demonstrate competence understanding and wrangling geospatial data is important. You can layer more fancy aspects on top of that. You can give a very good impression by being able to use geopandas’ .explore method or adding static basemaps using contextily, being able to plot a geometry boundary on top of an xarray DataArray plot etc.

Depending on the size of organization you end up joining, I’ve found that geospatial data science can put even more demands than usual on the data engineering aspect of the job, precisely because you often end up having to wrangle both vector and raster data and also worry about how it all will scale as well (so put some time into using dask to process chunked data).

Just some random suggestions for you - sorry I have to dash into a meeting now. Get back if you have any more questions!

1 Like

Thank you @Guy_Maskall for your response. So maybe I can explain my background. I graduated with a Bachelor of Science in Geomatics and Geospatial Information Systems. It’s a five-year course with a focus on GIS and Remote Sensing. Therefore my geospatial Background is quite solid. I have also pursued geospatial data science courses and workbooks.
Therefore my interest in this field is drawn from my background. I definitely want to be part of projects that deal with ecology, epidemiology, transport and e.t.c. Since I have some of the skills stated, might you know organizations that offer Internship or volunteer opportunities to help me gain some experience.

I believe contributing to open source is an excellent way to offer experience (and potentially get a job)!

Check out this doc about contributing:

Or feel free to ask here!

1 Like

Thanks @ahuang11, I’ll definitely look at the project.

Some more geospatial career advice that might be helpful - Start your Geospatial Career. This article shares some guidance and… | by Christoph Rieke | Aug, 2023 | Medium.