Skip to content
Lamiak

Lamiak Core

The system underneath every LAMIAK product.

Lamiak Core is the layer under every agent: it assembles context, routes each step, holds actions for approval, and measures what happened. Parts of it are built and tested today; every part below carries its current status.

The loop

Understand the work.Act with control. Learn from outcomes.

The loop in the abstract, then each of its stages with the status it carries in the registry today. No stage is offered to customers yet, and each product page shows its own status.

  1. 01 Context

    What the system knows: records, documents, permissions, history.

  2. 02 Action

    What an agent or workflow does, inside the rules you set.

  3. 03 Outcome

    What actually happened, recorded with its evidence.

  4. 04 Learning

    What changes next time, promoted only after review.

Context, then Action, then Outcome, then Learning; learning feeds back into context.

The stages, today and next

Today

  1. Context

    In development

    Gather the records and permissions a task needs into one packet before an agent acts.

  2. Approval

    Human approvalIn development

    Hold selected actions for a person, and record who decided and when.

  3. Routing

    In pilot

    Send each step to the right model for the task and the sensitivity of the data.

  4. Evaluation

    In development

    Score agents against suites and graders before trusting a change.

  5. Feedback

    In development

    Capture edits, rejections and outcomes from the people doing the work.

Next

  1. Dataset snapshots

    Planned

    Freeze feedback and outcome data into versions a decision can cite.

  2. Model registry

    Planned

    One identity for every model, adapter and prompt, with its evaluation history.

  3. Promotion gates

    Planned

    A candidate proves better on evaluation before it is given authority.

  4. Training from outcomes

    Planned

    Adapt models on outcome data a customer has agreed to share.

Next is where the loop closes: outcomes become data, data becomes a candidate, and a candidate earns its place on evaluation before it is given authority.

Today: context, approval, routing, evaluation and feedback. Next: dataset snapshots, model registry, promotion gates and training from outcomes. Learning feeds back into context.

The complete list, including the parts that are not loop stages, is on the Trust page.

Start with one workflow that matters.

We map the work, the systems it touches and the approvals it needs before anything is automated.