Enterprise Support · Groundwork Analytics
petropt is free, MIT-licensed, and stays that way. Groundwork Analytics is the team behind it — for the work that needs custom models, your data, and your basin.
No newsletter. No drip campaign. A real conversation about a real problem.
If any of these sound like the week you just had, the call will be useful.
Mismatched well names, gaps, allocations that don’t reconcile, units that drift between vendors. Engineers spend a week before any analysis can start.
Talk it through →
Decline curves that diverge between engineers. P50s that move with whoever ran them. Reserves auditors who keep flagging the same wells.
Talk it through →
Standing PVT was fit to 1940s Gulf Coast crudes. Your Wolfcamp B fluid doesn’t look like that. You need a correlation calibrated to your fluid system, not a textbook average.
Talk it through →
An honest line. So you know what you’re paying for — and what you’re not.
Free · petropt v0.3.1 Beta · MIT
Free for commercial work. No support contract required. GitHub →
Paid · Groundwork Analytics
Engagement-based. NDA on request. Scoped from a 30-minute conversation.
Six engagement types. We’ll tell you on the call which one fits.
Beyond textbook formulas: Bayesian decline-curve analysis with MCMC P10 / P50 / P90, multi-well anomaly detection, fluid recommendation from compositional databases, completion design optimization.
Turn the generic petropt.correlations into a tuned model for your basin, formation, and fluid system using your historical data.
Automated LAS QC, multi-well log analysis, real-time production surveillance. The data engineering layer that has to exist before any model runs.
Get your engineering team productive with petropt and Python in days, not months. Run on your data. Your engineers, not seat licenses, do the work.
Decline forecasts, type curves, material balance, rate transient analysis, EUR / 3P reserves work — either as a peer review of your team’s output or independent execution.
Pay to add a feature to petropt that benefits your team and the wider community. Your name in the changelog. The code stays MIT-licensed.
You describe the problem in your own words. We tell you whether petropt + your team can solve it free, or whether something paid is the right fit. Either answer is fine.
If there’s a fit, we sign an NDA, look at a representative slice of your data, and define a deliverable in writing — what gets built, what success looks like, who owns the IP.
Fixed scope, fixed price where possible. Code, models, and documentation handed back to your team so the work outlives the engagement. No vendor lock-in.
Former Schlumberger / Stanford ERE researcher and SPE author, and the maintainer behind petropt. petropt is the open-source layer of a broader analytics stack used in production by oil & gas operators.
Mutual NDA before any data review. Work runs in your cloud or on dedicated infrastructure for the engagement. For consulting work, data handling is governed by the SOW or a separate Data Processing Agreement, with terms agreed mutually before any data is shared.
Engagements are sized for independent E&P operators and PE-backed teams — the 200-to-5,000-well operator who needs custom modeling without enterprise platform pricing.
Yes. MIT license, public repo, public PyPI package. The free tier doesn’t shrink as paid offerings grow — if anything, it expands. Sponsored features go back into the public package.
Yes. Mutual NDA before any technical conversation that touches your data. For data processing, we work to your DPA template or a standard one we provide. We can work through your security review questionnaire as part of scoping.
IP and licensing terms are written into each SOW. Typical scoping starts from: you own deliverables built on your data; improvements to the open-source petropt library remain MIT-licensed and benefit the wider community. The exact split is agreed in writing per engagement — no ambiguity, no surprise carve-outs.
Independent E&P operators and PE-backed teams from a few hundred to a few thousand wells. We’re not the right fit for supermajors who already have an internal data-science org. We’re built for the operator who wants the model without the platform.
Typical cadence: intro call within a week or two of first contact, NDA shortly after if there’s a fit, then scoping. Real-world timing depends on your team’s availability for data review and on how clearly the problem is already defined.
A 30-minute call. Honest read. Either we can help, or we’ll point you to the open-source path that already does.
Already using petropt? Tell us what you’re building — we like the stories.