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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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
Read the post

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.