These are the engagements where the work and the result can be discussed publicly. Each one is a documented application of the ORCA framework on a real dataset — what we ran, what worked, what did not, and what the uncertainty estimates actually said.
A rare combination — sparse 2D lines and a real 3D survey acquired over the same area — used as the one-time opportunity to validate against true ground truth. Four horizons reconstructed, RBF and GP compared head-to-head, calibration measured against the 68% target. Results are mixed in an informative way: the method clearly wins on the well-behaved horizons and exposes a known 2D-vs-3D processing mismatch on Baccaro.
The same reconstruction framework applied outside oil & gas — time-lapse change detection from sparse repeat acquisitions over an active glacier. Demonstrates that the geological prior transfers, with appropriate retraining, to surface-and-near-surface targets at very different length scales.
P10/P50/P90 structural-closure envelopes built directly from sparse legacy 2D — the regulator-facing use case the framework was designed for.