Force-Fit vs Auto-Fit: Why "AI" Decline Curves Overstate Your Reserves
There is a specific, well-documented way to make an oil or gas well look more valuable than it is, and it hides inside one innocent-looking
#Oil & Gas
A Day in the Life: The Reservoir Engineer
"Petroleum engineer" is one job title covering sub-disciplines that share a hallway and almost nothing else about their day. This is the one
A Day in the Life: The Drilling Engineer
Of all the petroleum sub-disciplines, drilling is the one whose mistakes call you at 2 a.m. The drilling engineer argues with a rock that&
A Day in the Life: The Production & Operations Engineer
Drilling delivers one well at a time. The production engineer inherits all of them at once — and has to keep a few hundred of them
A Day in the Life: The Completions Engineer
Drilling makes the hole. Reservoir engineering says what's worth producing. The completions engineer stands in between and answers the question that makes or
A Day in the Life: The Reservoir Simulation Engineer
Reservoir engineering forecasts the asset. Reservoir simulation builds the physics model the forecast is argued against. They sound like the same job — outsiders, and plenty
The Missing Middle: What's Left in Oil & Gas When Private Equity Walks Away
There used to be a middle in American oil and gas. A working operator could raise twenty or fifty million dollars, lease up a few
Why Oil & Gas Operators Build Their Own Databricks
Upstream operators and energy investors run seismic, production, and sensor data on Databricks-shaped platforms billed by the second. A self-hosted open-source lakehouse is a real money and sovereignty decision for oil & gas — here's the honest version, with no invented savings numbers.
Where Decline-Curve Analysis Fits — and Why We Built a Browser-Native Engine for It
Decline-curve analysis is the load-bearing first step of almost every reserves number. We rebuilt the hard part to run entirely in the browser — matching Equinor's open-source math to 0.02 ppm, with no server and no data upload.
The open-source reservoir engineering stack — reference architecture for the bottom of the market
Three modules, sharing a Postgres backbone: ingestion, model/forecast, reporting+audit. Buildable in 4-8 weeks on $40-80/month infrastructure. The open-source replacement for the enterprise reserves stack at the bottom of the market.
What super-majors actually built — Johan Sverdrup and SLB's agentic AI reservoir architecture
Two recent reference architectures from the published literature: Equinor's Johan Sverdrup digital twin and SLB+ONGC's 2025 agentic AI framework for large-scale well modeling. Both describe the same architectural pattern.
The reserves engineering gap — three workflow gaps the 2024-26 literature is calling out
Small E&P operators and family offices are running 2026 capital on 1990s workflows. The 2024-26 reservoir engineering literature has converged on the diagnosis: the gap is in workflow infrastructure, not algorithms.
Mineral royalty cash flow to PV-10 — the valuation modeling step
Stage 5 of the Texas mineral valuation pipeline: the four inputs (Arps decline, price deck, lease economics, discount rate), the monthly cash-flow math, and the 13-line validation checklist that catches the recurring NPV errors.
Texas decimal interest math — pooled vs allocation walkthrough
Same well, two methods, 32% gap. The math behind why allocation-well decimal interest differs from pooled approximation, with a worked example and the PostGIS geospatial step that makes the answer defensible.
Surface parcels to net mineral acres — the data layer every Texas mineral valuation depends on
Stages 1 and 2 of a defensible Texas mineral valuation: acreage technique (43,560 sq ft, geodesic projection, source tiering) and ownership resolution (severed estates, UMI math, the 3/8 example). With an 8-line audit checklist.