Macro photograph of drill cuttings
Frontier AI for oil and gas

The well already knows.

The intelligence layer for drilling operations: agentic AI built on frontier models of the subsurface, in production across operators' programs today.

The problem

Data was never the bottleneck. Context is.

Rigs stream terabytes. Labs return reports. Dashboards multiply. The answer is already in there, scattered across forty-three systems that don't talk to each other, or to the people deciding what happens next.

Field engineer texting Derrick from the rig
Meet Derrick

Intelligence you can text.

Derrick lives where the work happens: iMessage, SMS, email, a voice memo from the truck. Ask about any well and he plans the analysis, queries the data, generates the plot, and answers with citations. And when action is on the table, a comment in StarSteer, a push to WITS, he asks first.

Derrick is not a feature. He is a contact in your phone, a canvas on your screen, and a memory that compounds with every well drilled.

PDC drill bit portrait
PDC bit · 8¾ in
Grounded

It starts at the bit.

Software-only platforms stop at the sensor feed. DrillSense is grounded in the physical well: the dynamics, the chemistry, the rock itself as it comes up, and the judgment of the people closest to the bit. Intelligence with ground truth under it.

The posture

DrillSense doesn't replace your tools. It makes them intelligent.

Your data flows in, gains subsurface context, and flows back out: enriched, explained, and entirely yours. No lock-in. No black boxes. No trapped models.

Deeply integrated with the stack operators already run: the labs, the logging units, the geosteering platforms, the drilling software. Each claimed its layer. Engineering, geology, operations. DrillSense is the AI layer across all of them.

Drilling rig at blue hour
Already deployed

Not a pilot.

Running at critical decision points today, from supermajors to independents.

38
Active operators
4
Basins
8
U.S. states
The lab

The future of energy AI is not generic chat over documents. It's connected, contextual engineering intelligence that understands the well, the rock, the operation, and the lifecycle of decisions that follow.

Beyond the application layer, we're investing in proprietary subsurface AI.
01 Patent pendingSubsurface world modelA foundation model that learns the physics of the subsurface from drilling itself.
02 Patent pendingFederated learningLearning across operators without exposing anyone's data.
03GPU infrastructure at scale, ownedFrom B200 clusters to edge hardware on the rig. Our compute, edge to core.
04AI Safety & TransparencyObservability, trust and transparency as tenets. Every model explainable and governed. SOC 2 in progress.

The Human in the Loop

AI intelligence for the real world. Curated weekly by humans. Read past editions

Weekly. No noise. Unsubscribe anytime.
Abstract rock strata texture

See it on your wells.

DrillSense is rolling out with select operators across the well lifecycle: drilling, completions, and production operations. The fastest way to understand it is to put it to work.