AI Firm Ambyint’s New Bakken Deal With Equinor Moves the Industry Another Step Closer to the Edge
Bibliographic record
Abstract
In the face of leaner economic times, oil and gas companies want to be able to boost well counts while minimizing any new additions to their field workforce. Among the companies answering this call is Ambyint, which was just tapped to deploy its artificial intelligence (AI) systems to help optimize all of Equinor’s Bakken Shale wells running on sucker rod pumps—the oil field’s most common breed of artificial lift. The deal is understood to be one of Ambyint’s largest contracts to date. Specific numbers have not been shared, but public data show that Equinor operates more than 800 wells in North Dakota. Ambyint, which is headquartered in Calgary and has an office in Houston, said the project scope may eventually include all of the Norwegian-owned operator’s horizontal wells in the state as they transition to rod pump. The large-scale upgrade marks another milestone in the evolution of oilfield automation that for decades has been defined by a nearly-ubiquitous reliance on SCADA (supervisory control and data acquisition) systems. One of the drawbacks of the status quo is that it requires small armies of field personnel to interpret SCADA data and then adjust setpoints to get pumping units back into optimal operating ranges. This manual process can consume half an hour per well to complete; downtime that quickly adds up in a field of hundreds. “What we are talking about is having the machine do that entire workflow,” Chris Robart, Ambyint’s president of US operations said. By equipping wells with its cloud-based AI and edge computing technology, and an application the company calls an autonomous setpoint management system, the man-hours once spent re-setting pumps can hopefully be reallocated to other bottom-line drivers. “We are freeing up individuals to go do other things, like think about new technology, troubleshoot failed equipment, deal with workovers, or new well designs,” Robart added. Pilot Hits the Mark for Equinor The Bakken project comes after a pilot that included 50 of Equinor’s wells, which saw a net production increase of 6%—considerably larger uplift figures were seen from those wells suffering from underpumping. The encouraging results were realized with zero shutdowns in production and minimal “human interference,” according to the companies. “The Ambyint technology has improved the remote data visibility and has delivered a more accurate diagnostic of downhole conditions to our rod pump wells in the Bakken,” a production engineer for Equinor’s Bakken asset, Jack Freeman, said in a statement. “The autonomous speed range management tool has leveraged the power of machine learning to optimize our wells by identifying and acting on real opportunities” Last year, Ambyint’s Series A funding round raised $11.5 million from several venture capital groups including Equinor Technology Ventures (formerly known as Statoil Technology Invest).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".