The Cortland blog
Field notes from Claude in production.
What we’re learning as Claude moves from a chat window into the plants, pipelines, and portfolios our clients operate. Not a roundup. Not a recap. Notes from the work.
May 29, 2026
Why AI Deals Stall in Procurement, and How to Get Them Through
In large energy and industrial companies, AI projects rarely die in the demo. They die in procurement. Here's why enterprise vendor governance stops AI deals, and how the person inside who wants the project can actually get a good one through.
- pod-a
- procurement
- vendor-governance
- oil-and-gas
May 22, 2026
AI Governance in Process Safety: Working Inside PSM
Any AI that touches safety-critical refinery operations runs into PSM, MOC, and PHA review. That's not an obstacle to route around. It's the environment good industrial AI is designed for, and Cortland builds inside it.
- pod-a
- industrial-ai
- governance
- process-safety
May 15, 2026
What Your Houston Peers Are Actually Doing With AI
Supermajors benchmark each other obsessively on AI. The public scoreboard of partnerships and platforms tells you less than it looks like. Here's how to read your real position, from someone who isn't selling you a platform.
- pod-a
- industrial-ai
- benchmarking
- oil-and-gas
May 8, 2026
Capturing the Crew Change: AI for Institutional Knowledge
The most experienced reliability engineers in oil and gas are retiring, and decades of judgment are walking out with them. AI can capture what's leaving and equip who's arriving. Here's how the two halves fit together.
- pod-a
- industrial-ai
- workforce
- knowledge-transfer
May 1, 2026
The Pilot-to-Production Gap in Houston Supermajors
By most industry estimates, the large majority of supermajor AI pilots never reach production. The gap between a promising demo and a system operators actually rely on is where the value sits, and where Cortland does its work.
- pod-a
- industrial-ai
- scaling
- oil-and-gas
April 24, 2026
Running Production AI Without a Data-Science Team
A lean operator can deploy AI. Keeping it running is the part nobody staffs for. Here's how to keep production AI accountable, drift, cost, and ownership, with minimal headcount.
- pod-b
- independent-ep
- ai-operations
- oil-and-gas
April 20, 2026
Claude is the UI
Industrial operators aren't going to log into another dashboard. They're going to ask Claude. Here's what that changes for the software you already own.
- cadence
- positioning
- industrial-ai
April 17, 2026
Why the Model That Worked on One Pad Fails on the Next Basin
Upstream AI models that shine in one field often break in the next, and the model never warns you. Here's the generalization problem in E&P AI, and how to build for it from the start.
- pod-b
- independent-ep
- ai-build
- oil-and-gas
April 10, 2026
AI Literacy for Lean E&P Teams: Punch Above Your Headcount
A small operator can't hire a data-science team, and doesn't need to. Here's what AI literacy actually consists of for a lean E&P team, and how to build it without adding a single role.
- pod-b
- independent-ep
- ai-literacy
- workforce
April 3, 2026
Which AI Use Case Actually Pays Back in an Independent E&P
For a capital-disciplined operator, the AI question isn't whether it can do something. It's what the payback is. Here's a practical way to pick the one use case worth doing, instead of chasing the demo that impressed someone.
- pod-b
- independent-ep
- ai-use-cases
- oil-and-gas
March 27, 2026
How Independent E&Ps Can Out-Adopt the Supermajors
The supermajors have the AI budgets and the internal platforms. Independents have something money can't buy back quickly, namely speed and fewer silos. Here's how a lean operator can out-adopt the majors before that gap closes.
- pod-b
- independent-ep
- ai-strategy
- oil-and-gas
March 20, 2026
AI Literacy for the Control Room and Integrity Desk
Operators and integrity engineers don't need to become data scientists. They need enough fluency to use AI well and to know when to override it. Here's what that fluency actually consists of.
- pod-c
- midstream
- ai-literacy
- workforce
March 13, 2026
When Your Leak-Detection AI Cries Wolf
A leak-detection model that floods the control room with false alarms gets ignored. One that misses a real leak is a safety event. Operating between those two is the real work, and it isn't a model-accuracy problem.
- pod-c
- midstream
- ai-operations
- leak-detection
March 6, 2026
Turning SCADA and Historian Data Into Decisions
Midstream operators already collect oceans of telemetry and act on a sliver of it. The opportunity isn't more sensors. It's making the data you already have answerable.
- pod-c
- midstream
- ai-build
- reliability
February 27, 2026
The Highest-Uptime AI Use Case in Midstream
Midstream earns on throughput, so the right way to judge an AI use case is how much uptime it protects per dollar. Here's how to choose between leak detection, compressor health, and integrity analysis without letting a vendor choose for you.
- pod-c
- midstream
- ai-use-cases
- reliability
February 20, 2026
Governing AI Inside PHMSA Integrity Management
Any AI that touches pipeline integrity enters a federally regulated, auditable environment. That isn't a reason to keep AI out of integrity work. It's the specification good integrity AI is built to meet.
- pod-c
- midstream
- governance
- pipeline-integrity
February 13, 2026
AI Literacy for Process and Reliability Engineers
Process and reliability engineers don't need to become data scientists. They need enough fluency to use AI well inside a safety-critical plant and to know when to override it. Here's what that fluency consists of.
- pod-d
- refining
- petchem
- ai-literacy
February 6, 2026
Operating AI in a Coupled Process
In a refinery, nothing is isolated. A model that drifts or nudges the wrong recommendation sends ripples through a coupled system. Operating AI here is less about the model and more about the discipline around it.
- pod-d
- refining
- petchem
- ai-operations
January 30, 2026
Putting AI on Top of the DCS Without Touching the Control Loop
The fastest way to get AI into a plant is to put it next to the control system, not inside it. Advisory intelligence on top of the DCS and historian delivers value without crossing the safety boundary.
- pod-d
- refining
- petchem
- ai-build
January 23, 2026
The Highest-Margin AI Use Case in a Refinery (That Can Actually Pass MOC)
In refining and petchem, margin lives in the process. Yield, energy, throughput. The best AI use case is the one that improves a margin lever and can still clear Management of Change. Here's how to find the overlap.
- pod-d
- refining
- petchem
- ai-use-cases
January 16, 2026
MOC for AI: Deploying Models Under Management of Change
Any AI that touches a process unit runs through Management of Change before it goes live. That isn't bureaucracy to route around. It's the discipline that makes AI safe to deploy in a plant, and building for it from the start is what separates models that ship from models that stall.
- pod-d
- refining
- petchem
- governance
January 9, 2026
AI Literacy for Grid Operators, Plant Engineers, and Market Analysts
The people running the grid, the generation fleet, and the market desk don't need to become data scientists. They need enough fluency to use AI well in a critical-infrastructure setting and to know when to override it.
- pod-e
- power
- utility
- ai-literacy
January 2, 2026
Operating AI When an Outage Is a Public Event
In most industries a model error is an internal problem. In power, a wrong forecast or dispatch call can show up as an outage on the evening news. That raises the bar on how AI gets operated after it launches.
- pod-e
- power
- utility
- ai-operations
December 26, 2025
Putting AI on the Grid Data You Already Have
Utilities and generators sit on enormous operational data, AMI, EMS, SCADA, generation telemetry, and use a fraction of it. The opportunity isn't more sensors. It's making the data you already have answerable, without touching the control path.
- pod-e
- power
- utility
- ai-build
December 19, 2025
Where AI Protects the Most Reliability in a Power Business
In a power business, reliability is the product, and an outage is a public event. The best AI use case is the one that protects the most reliability per dollar. Here's how to find it across forecasting, outage prediction, dispatch, and equipment health.
- pod-e
- power
- utility
- ai-use-cases
December 12, 2025
Deploying AI Inside NERC CIP and the ERCOT Market
Any AI that touches the grid or the generation fleet lands inside two regulated worlds at once. NERC CIP governs the cybersecurity of the bulk-power system, and ERCOT and FERC govern market behavior. Building for both from the start is what separates AI that ships from AI that stalls.
- pod-e
- power
- utility
- governance
December 5, 2025
AI Literacy for Engineering and Delivery Teams
In a services business, your engineers' time is the product. AI literacy isn't a nice-to-have for them; it's a margin and a credibility issue. Here's what that fluency actually consists of.
- pod-f
- energy-services
- engineering
- ai-literacy
November 28, 2025
When Your AI Has to Clear Your Client's Governance
Scaling an AI-enabled offering across clients means scaling across their governance, not just your own. A product that sails through one operator's review can stall at the next. Building for that is the difference between an offering and a one-off.
- pod-f
- energy-services
- engineering
- ai-operations
November 21, 2025
Turning a Service You Deliver by Hand Into a Product
The most valuable AI move in a services business isn't doing the work faster. It's turning a service you deliver by hand into a repeatable product. That crossing is harder than it looks, and it's where the lasting value is.
- pod-f
- energy-services
- engineering
- ai-build
November 14, 2025
Internal Efficiency or a New Product? Where AI Pays Off in a Services Business
AI can make a services firm deliver the same work cheaper, or it can turn that work into a new product. Those are different bets with different payoffs. Here's how to choose which one to make first.
- pod-f
- energy-services
- engineering
- ai-use-cases
November 7, 2025
When Your AI Becomes Your Client's AI
When you build AI into what you deliver, it doesn't live under your governance. It lives under your client's. For an energy-services or engineering firm, that changes what good AI even looks like.
- pod-f
- energy-services
- engineering
- governance
Have a production problem Claude should be solving?
The work on the blog is the work we do in engagements. If one of the posts lines up with something you’re wrestling with, let’s scope it.