Balaguroff

Balaguroff

From the first question
to the exploration well that proves it

From the first question to the exploration well that proves it

I bring data of every scale into one model. Algorithms do the routine checks. The decision and the signature stay with the geologist.

QuestionDataModelDecisionProof
8 tools and projects. 4 of them are open-access tools with a DOI.

Maxim Balagurov, petroleum geologist

One question runs through everything here: what does the subsurface hold, and how sure is the answer? Balaguroff is the name of my line of tools and projects.

I take a subsurface question from the first idea to the drilling decision. I define the question, build the team and take responsibility for the recommendation.

I take a subsurface question from the first idea to the drilling decision. I define the question, build the team and take responsibility for the recommendation.

I integrate seismic, well, gravity-magnetic, core, laboratory and field data into geological models and well recommendations.

At CGG, I recommended 30+ wells across 15+ exploration areas; every well drilled since has confirmed the predicted targets.

I integrate seismic, well, gravity-magnetic, core, laboratory and field data into geological models and well recommendations.

At CGG, I recommended 30+ wells across 15+ exploration areas; every well drilled since has confirmed the predicted targets.

One question runs through everything here: what does the subsurface hold, and how sure is the answer? Balaguroff is the name of my line of tools and projects.

I take a subsurface question from the first idea to the drilling decision. I define the question, build the team and take responsibility for the recommendation.

I integrate seismic, well, gravity-magnetic, core, laboratory and field data into geological models and well recommendations.

At CGG, I recommended 30+ wells across 15+ exploration areas; every well drilled since has confirmed the predicted targets.

I integrate seismic, well, gravity-magnetic, core, laboratory and field data into geological models and well recommendations.

At CGG, I recommended 30+ wells across 15+ exploration areas; every well drilled since has confirmed the predicted targets.

Selected Work

Selected Work

The same questions, in real projects.

The same questions, in real projects.

Selected Work

The same questions, in real projects.

The next step of the same work

When the expert retires, the knowledge stays

The same archive, one step further: it answers questions and explains itself on video.

My role

I work closely with IT developers, some of them with oil-company experience.

Illustrative clip: the same threads, first tangled, then in order.

Answers from a company’s own library

A search system for a company’s reports, methods, standards, maps and models. It answers only from these documents and cites the page. It runs inside the company’s own network.

Pilot stage · open to partners

Documents turned into video

A PDF or DOCX becomes a narrated video with the document’s own figures on screen, in 175 languages. It fits tutorials, training and the presentation of results.

Already in use · built with a partner team

Tools and projects

All tools and projects in one list. Each one shows its stage of development. The open-access tools share one DOI: 10.5281/zenodo.24242389.

4 open-access tools · 1 in a project · 3 next steps

Open-access tools · DOI

01

Proof

Prediction vs. result

How close were the predictions to what the wells found?

Based on: my recommendations for wells at CGG →
Open-access tool · DOI
02

Data

What must the survey resolve?

A check of a planned magnetic survey before money is spent on field work.

Based on: What should the next geophysical survey be? →
Open-access tool · DOI
03

Data

Five wells: which five?

A plan for shallow stratigraphic drilling when the wells are few.

Based on: Which target should be drilled first? →
Open-access tool · DOI
04

Decision

How much, and how sure?

The oil volume of an exploration prospect, as a range from P90 to P10, and the data that could change the decision.

Based on: Which target should be drilled first? →
Open-access tool · DOI

In a project

05

Data

Digital Geology: Cross-Checking Historical Well Data

AI agents read the archive; conflicts are flagged for a geologist.

Pilot on a producing-field project

Next steps

06

Question

A company library that answers

The next step is a pilot on a partner's own documents.

Based on: What is already in the archive? →
Pilot stage · open to partners
07

Decision

Documents turned into video

Works today for tutorials, training and presenting results.

Based on: What is already in the archive? →
Already in use · built with a partner team
08

Data

UAV survey as the next data layer

The programme was proposed in 2025. A licensed operator makes the survey flights.

Based on: What should the next geophysical survey be? →
Proposed programme

Where access is hard, a UAV survey can bring the data.

Where access is hard, a UAV survey can bring the data.

Tundra, ice and remote outcrops are hard for a geologist to reach. UAV surveys can bring them into the model. This is a proposed programme.

Tundra, ice and remote outcrops are hard for a geologist to reach. UAV surveys can bring them into the model. This is a proposed programme.

Illustrative view

Illustrative view

How a project runs.

How a project runs.

UAV lifting off a rocky northern plateau at low sun

1

Data

I bring seismic, wells, earlier interpretations and new field surveys into one checked data set. Conflicts between datasets are logged and resolved before modelling. Output: one checked data set and a list of assumptions.

UAV lifting off a rocky northern plateau at low sun

1

Data

I bring seismic, wells, earlier interpretations and new field surveys into one checked data set. Conflicts between datasets are logged and resolved before modelling. Output: one checked data set and a list of assumptions.

UAV lifting off a rocky northern plateau at low sun

1

Data

I bring seismic, wells, earlier interpretations and new field surveys into one checked data set. Conflicts between datasets are logged and resolved before modelling. Output: one checked data set and a list of assumptions.

Framed 3D block model of a mountain with faulted layers below it

2

Model

I build the faults, layers and hydrocarbon traps on checked data. Other possible models are kept as scenarios, and the uncertainty is mapped. Output: a structural model with scenarios and an uncertainty map.

Framed 3D block model of a mountain with faulted layers below it

2

Model

I build the faults, layers and hydrocarbon traps on checked data. Other possible models are kept as scenarios, and the uncertainty is mapped. Output: a structural model with scenarios and an uncertainty map.

Framed 3D block model of a mountain with faulted layers below it

2

Model

I build the faults, layers and hydrocarbon traps on checked data. Other possible models are kept as scenarios, and the uncertainty is mapped. Output: a structural model with scenarios and an uncertainty map.

Empty high-floor room with a city skyline behind the glass

3

Decision

The result is a clear recommendation: • Go / Conditional Go / Redesign / Stop • Zones where drilling is safe enough • Uncertainty map and a list of assumptions • A short summary for managers

Empty high-floor room with a city skyline behind the glass

3

Decision

The result is a clear recommendation: • Go / Conditional Go / Redesign / Stop • Zones where drilling is safe enough • Uncertainty map and a list of assumptions • A short summary for managers

Empty high-floor room with a city skyline behind the glass

3

Decision

The result is a clear recommendation: • Go / Conditional Go / Redesign / Stop • Zones where drilling is safe enough • Uncertainty map and a list of assumptions • A short summary for managers

Who takes part
The team is assembled for each question.
I lead the geology and the decision. Specialists and partners join at the stage where their contribution changes the result.
Question
Data
Model
Decision
Proof
Project lead
Operator's geoscience team
Geologists and geophysicists
Laboratories
Data and AI engineers
Research groups and institutes
Data processing and geophysical inversion
Economists
Licensed operators, drillers
Who takes part
The team is assembled for each question.
I lead the geology and the decision. Specialists and partners join at the stage where their contribution changes the result.
Project lead: at every step
1
Question
Operator's geoscience team
2
Data
Operator's geoscience team · Geologists and geophysicists · Laboratories · Data and AI engineers
3
Model
Geologists and geophysicists · Data and AI engineers · Research groups and institutes · Data processing and geophysical inversion
4
Decision
Operator's geoscience team · Economists
5
Proof
Data processing and geophysical inversion · Licensed operators, drillers
Who takes part
The team is assembled for each question.
I lead the geology and the decision. Specialists and partners join at the stage where their contribution changes the result.
Project lead: at every step
1
Question
Operator's geoscience team
2
Data
Operator's geoscience team · Geologists and geophysicists · Laboratories · Data and AI engineers
3
Model
Geologists and geophysicists · Data and AI engineers · Research groups and institutes · Data processing and geophysical inversion
4
Decision
Operator's geoscience team · Economists
5
Proof
Data processing and geophysical inversion · Licensed operators, drillers
Biplane lifting off a dry lake bed with a man standing beside it. Before it flies, someone stands beside it.
Biplane lifting off a dry lake bed with a man standing beside it. Before it flies, someone stands beside it.

Questions about a job, a project or joint research? Write to me.

Questions about a job, a project or joint research? Write to me.

Questions about a job, a project or joint research? Write to me.

No confidential data is needed for a first conversation.

No confidential data is needed for a first conversation.

Start a conversation