Synthetic seismic section: an anticline trap with a gas bright spot and a flat spot

Data liberation · Restructuring · Evaluation

Liberating Indonesia’s subsurface data.

Most oil and gas data is still on tape and paper. ASTEC brings it back, structures it, and puts it to work, so your G&G teams can find it, trust it and use it. Then it is ready for AI.

Synthetic section: anticline gas trap, no client data

Everyone is talking about AI. The data is still in boxes.

Decades of seismic, well logs and reports sit on tape and paper that fewer and fewer machines can read. Before any of it can be searched, evaluated or used to train a model, it has to be liberated. That is the foundation we build, one archive at a time.

The data journey

From physical data to AI, one step at a time.

All services →
  1. Physical data

    Reels, cartridges, paper logs and reports in a warehouse.

  2. Data liberation

    Every block read off the tape, every page scanned.

    • Available now Data transcription
    • Coming soon Scanning and digitizing
  3. Data restructuring

    Catalogued and reorganized so data is easy to find and load.

    • Available now PPDM-aligned cataloguing
    • Coming soon Data restructuring
  4. Data evaluation

    Exploration and exploitation studies by our G&G team.

    • Available now G&G studies
  5. AI / ML

    Search, classify, extract and connect the archive.

    • In development AI and machine learning
ACROSS ALL STAGES

Data management: One organized home for your subsurface archive, looked after over time.

Coming soon

Why ASTEC

Built by the people who use the data.

ASTEC is not an IT company that learned oil and gas. Behind it is a G&G team that knows how subsurface data flows through a project, how it should be loaded, and where it hurts. We have waited weeks for data ourselves, and we know what changes once it is liberated and structured.

About ASTEC →

We know the workflow

Data is delivered the way an interpretation project expects it, not just as a folder of files.

We know the pain points

Surveys nobody can find, tapes nobody can read, files nobody can load.

We evaluate it too

The team that liberates your data can also run exploration and exploitation studies on it.

We want to find more

We are eager to see this data lead to the next hydrocarbon potential.

10+ tape formats

Media we read

See all media and capacities →
  • 9-track

    up to 180 MB native

    Field and processed seismic, well logs.

  • 3490E

    800 MB native

    Field and processed seismic, log archives.

  • 3590 / 3590E

    10 to 20 GB native

    Field data from larger 2D and 3D surveys.

  • 3592

    up to 7 TB native

    Modern 3D acquisition and processing archives.

  • 8mm Exabyte

    up to 5 GB native

    Field data, well logs, workstation backups.

  • SDLT

    up to 160 GB native

    Processing outputs and project archives.

Why it matters now

Aligned with Indonesia’s oil and gas data regulations.

Indonesia’s rules on oil and gas data push for data that is easier to hand over, verify and access. Liberated, structured data is how you get there.

Operators and data managers

Archives that can be handed over, verified and found, instead of boxes on a shelf.

G&G teams

Data that loads cleanly, so evaluation is faster and more accurate.

Investors

Data that is easier to access and assess when new acreage is on the table.

AI / ML · In development

AI is the last step of the journey, not the first.

We are building workflows to search, classify, extract and connect across your archive. They start from data we have already liberated and structured, so you can trust what comes out.

Join the AI pilot list

What is on your shelf?

Send a media list or a photo of a label. We reply with a reading plan, a timeline and a price.

Contact us