We offer a wide range of products and consulting services built upon traditional methods and AI (i.e., machine learning). If you have a problem that overlaps with our domain expertise in geostatistics, hydrogeophysics, data processing, modelling and simulation, or if you need a custom machine learning-based solution, we can help.
Learn more
Dewar AI grew out of many conversations between friends completing their PhDs at Stanford University over the many AI-related opportunities in the earth sciences. Over time we have coalesced into a diverse team of subject matter experts and collaborators united by a common goal: tackle challenging problems and make the world a better place.
Learn moreMany of the solutions and consulting services we offer are grounded in state-of-the-art research. Below is an assortment of the research that inspired some of our offerings.
Innovative method developed as part of Noah Dewar's PhD research that allows for the estimation of the depth to the water table from airborne electromagnetic data.
Learn more
Research proof of concept that led us to develop our automatic geophysical data processing workflow.
Learn more
This paper contains convolutional neural network architectures that inspired some of the models we use for time-series forecasting.
Learn more
Example of a graph convultional neural network that allows for temporal and spatial variation with a connected graph and Long-Short Term Memory structure; this structure is useful for problems that involve spatial and temporally varying data.
Learn more