A shot at perfection

Build the most profitable agent in a complex world

Complete the volume within a volume.

AI at the speed of sound

Segment and classify kerogen macerals in reflected light polished sections

Find unique patterns in time series data

Help Scrooge save on their Holiday electricity bill

Unsupervised Segmentation of thin section images

If it glows, we can model it

Robust prediction in unpredictable systems

Discover new demand response frontiers

Register the points on a crossplot

Identify horizons in seismic data

The first of three challenges to investigate the future of electrification

Progressing the “From One to Many” challenge

Erode the ordinary, and uncover the extraordinary

Find small features on satellite imagery

Customize a Generative AI model

Create a flexible model that can characterize steam growth in reservoirs using data from the South Belridge Field

The goal for this challenge is to find the real world coordinates (i.e. georeference) for map images.

This challenge will have you use rock radioactivity measurements to identify which body of water created them.

Mapping faults in the subsurface

Denoise subsurface seismic images

A seismic jigsaw puzzle

Restoring a seismic image one vertical slice at a time

Classify buildings based on their load profile curves

This challenge seeks new and fun ways to visualize geoscience data The sky's the limit on this competition; you can pick from several data types and any language to build a novel way to explore the data. The more imaginative you get, the better!

Generate an agent-based model that mimics fluid movement in the subsurface

Using one scientific discipline to help another

This challenge will help you learn how to use Python to work with geoscience data. The tools will be simple to understand and research but have practicalities with real data and provide a foundation for further learning and growth.

Create a machine learning model that has the highest accuracy in prediction lithology from a suite of wireline logs. A training dataset with hand interpreted and QC’ed wellbore lithology is available

Builds the best machine learning based fault mapping algorithm for seismic data