Skip to main content

Insights

Sharing how the people building the field see the future of energy.

Featured
AI

The 288-Slide Running Log: Weekly Evidence Beats Monthly Polish in Applied-AI Delivery

On a roughly year-long applied-AI engagement with a major operator in Oman, the artifact we trusted most was not the monthly steering deck. It was a single running-log deck that grew by appending every week: 288 extracted text blocks holding raw ground-truth-versus-prediction tables, a NaN-imputation status ledger that named its own failures (GAN and GAIN Fail, the masked autoencoder still Optimizing, KNN Succeeded with spikes), and the all-caps escalation the day the model hit a wall on missing data. This is an operating-model piece about why that living log, not the polished summary, is the real system of record, and how it later became raw material for two journal papers.

Blog Article|7 min read
Showing 7 of 382 insights
Case study/Subsurface AI

The Attrition Ledger: How We Ran a Subsurface-AI Program on One Number, Asked-Versus-Delivered

Why the well count and its intake gate were the real project-management artifact of a three-phase carbonate borehole-imaging program in Oman. The running tally read 5-in / 0-kept, then 10-in / 8-kept, then 25-asked / 8-delivered, and we tracked the asked-versus-delivered gap every month on a data-management dashboard. This is a first-person account of the ledger discipline itself: how we ran the program on one number so that, by the final report, no one was surprised by the 8.
Tannistha MaitiTannistha MaitiSenior AI Researcher
Case study|7 min read
Case study/Subsurface AI

The Pilot That Pivoted: How a Geomechanics Proposal Shipped as a Fracture-Detection Transformer

The engagement we were funded to run in August 2020 was a 10-month geomechanics ANN estimating in-situ stress and elastic moduli from borehole deformation. What shipped, three years later, was a fracture, bedding and vug detection transformer. This is the honest account of how the technical core pivoted without the contract dying, and why the phase-gate structure is what made that survivable.
Tannistha MaitiTannistha MaitiSenior AI Researcher
Case study|8 min read
Case study/MLOps

Agentic MLOps: From 6-week retrains to overnight

A global supermajor cut model retrain cycle time 40× and lifted production-forecast accuracy 22% using an agentic MLOps loop that monitors drift, reconciles data, and retrains autonomously — with geoscientists as approvers, not operators.
Tarry SinghTarry SinghFounder & CEO
Case study|8 min read
Case study/Computer Vision

From Pixels to Pores: Automated Vug Detection in Carbonate Reservoirs

A computer vision pipeline achieved 2,000× finer granularity than manual interpretation, recovering vugs missed by experts and delivering per-vug geometry at 0.1 m resolution across 200+ m of Middle East carbonate borehole image logs.
Tannistha MaitiTannistha MaitiSenior AI Researcher
Case study|7 min read
Case study/Computer Vision

Deep Learning for Fracture Detection in Borehole Images

A DETR-based model achieved 85% sensitivity for fracture detection in Middle East carbonate plays, cutting manual interpretation to validator-ready output.
Tannistha MaitiTannistha MaitiSenior AI Researcher
Case study|7 min read
Stay ahead

EarthScan insights, in your inbox.

Field-tested research on subsurface and energy-transition AI. About twice a month. No noise.

We use your email only for this newsletter. Unsubscribe anytime Privacy.