Records that never connect
Site conditions, sample results, and operating history sit in paper logs, spreadsheets, and individual memory.
NRS Labs Inc. · Alberta, Canada
Field data into operating decisions.
NRS Labs is building FieldSense, an AI operations platform for the oil and gas industry that turns field, laboratory, and sensor data into day-to-day operating decisions.
FieldSense · the platformTwo cores · engine and foundation modelChemPilot · the first module
01 The problem
Operations still run on paper, spreadsheets, and personal experience.
Decisions are made at every stage of the day, and they rest largely on manual work and individual judgment.
ProductionTreatmentWater handlingMeasurement Chemical treatment
Chemical treatment is one decision in that run, and it is where FieldSense starts.
A shrinking bench
Decisions depend on senior experts whose knowledge leaves the industry when they retire.
Found out late
What performed well last month can quietly fall out of specification as conditions drift, and the operator often learns of the problem late.
02 The platform
FieldSense is the product. Everything else is a layer of it.
FieldSense brings that information into one digital system, places AI recommendations in front of the people who make operating decisions, and learns from every result.
Product
FieldSense Role-based operating interface, cross-platform
Modules
ChemPilot First module
Further modules
Cores
Hybrid Field Engine Simulation and data generation
Field Foundation Model Continuously trained base model
FieldSense CaptureRecommendOperator decidesMonitorLearn
The platform keeps humans in the loop by design, and operators retain final authority throughout. The system recommends and explains, people decide.
03 The two cores
Two cores that build each other.
The decisive knowledge lives largely in expert experience rather than in public datasets. Neither side yields to off-the-shelf models.
Laboratory measurementReal field dataMultimodal sensing Hybrid Field Engine Field Foundation Model FieldSense Validated results return to the engine
Simulation and synthetic data generation
Hybrid Field Engine
Combines physical and chemical modeling with observation and sensing methods. Everything it produces is verified against independent laboratory measurements and field data from real site partners.
The continuously trained base model
Field Foundation Model
Built on the data the engine generates and verifies. Every new module, client, and validated result deepens it further.
04 The first module
ChemPilot is where the platform starts.
ChemPilot serves as the entry point for the platform. It carries early-stage validation, brings the first clients into the system, and opens the path for the same architecture to extend, direction by direction, into further modules.
A single chemical family may contain dozens of candidate products, and because chemicals are commonly applied in combination, the practical search space grows multiplicatively.
Families Candidate products, dosages, combinations
010203 Results feed the next recommendation
Digitalize Site conditions, sample properties, and chemicals on hand, in one standard form.
Recommend Candidate chemicals, combinations, dosages, and operating conditions, with the expected outcome.
Monitor Tracks performance, warns early when results drift, and updates the recommendation.
Engineers review and approve every recommendation, and each laboratory and field result feeds back into the system.
05 R&D force
Three organizations, one target.
NRS Labs Inc. is incorporated in Edmonton, Alberta, and is at an early building stage with research and development resources already in place.
ROSS Group, Computing ScienceInterfacial science laboratoryFlywheel ResourcesIn-house experimentation pipeline Expert-level decision quality at lower cost
University of Alberta
ROSS Group
Remote observation, smart sensing, and intelligent processing.
ross-research.ca ↗University of Alberta
Interfacial science laboratory
The science that governs how production chemicals work, with records accumulated over years of prior work.
ualberta.ca ↗Clearwater trend, Alberta
Flywheel Resources
A producing Alberta operator. Contributes real operating needs and historical field data for validation, testing, and training.
flywheelresources.ca ↗Most early AI companies begin with public data and search for a problem. NRS Labs begins with a documented industry pain point, deep domain expertise, an in-house research pipeline, committed data access, and a partner prepared to test in the field.
06 What comes next
Build, validate, then extend.
- Current
Build ChemPilot
Digitalize historical screening records and test recommendations against expert decisions.
- Next
Guided field trials
Run the recommendation and operator-review loop with our industry partner.
- Then
Module by module
FieldSense extends into related workflows that reuse the same loop.
- Compounding
The cores deepen
Every module and client feeds the engine and the foundation model.
07 Market
A continuing obligation, not a one-time study.
ChemPilot provides measurable benefits against that baseline, reducing physical screening tests, decision time, chemical consumption, and off-spec incidents.
Figures are third-party estimates, dated as published and not independently verified by NRS Labs.