Bayram Eker
EN
CV Start a conversation

Home / Writing

Writing: Systems

How AI, robotics and software fit together, including the decisions behind NODERIQ and QFlow.

Choose a topic and a format to narrow the archive.

Clear filters
Program state enters a decision model; a typed result and its uncertainty pass to a policy in code before an action is taken.

When software needs a decision, not another paragraph

A billing message needs the right destination. TypeSafe’s Jev offers one way to turn an interpretation into a result that software can act on.

Read
  1. A model proposes an action, the harness checks authority, and the environment returns evidence for the next step.

    When an AI agent says “done”, what should we check?

    A convincing completion message is only the beginning. Harness engineering is the work of helping an agent act, check the result and recover when needed.

  2. A wheeled inspection robot with a local compute enclosure stands beside industrial pipes; a service laptop and loose network cable sit nearby.

    The connection drops. The robot is still there.

    A closed door cuts a robot off from the cloud. Following one inspection explains why small models, local decisions and a shared memory of the job matter.

  3. Uncrewed assurance console showing decision evidence and authority state

    Why I keep the planner classical

    A quantum component can help with a calculation without taking over the decision. Keeping that distinction clear makes QANTIS easier to test and use.

  4. Enterprise evidence review inside QFlow Studio

    When does a quantum method deserve another experiment?

    A promising result leaves a practical question: should we keep going? Four checkpoints help us decide what to test before investing more time.

Also available as an RSS feed.

Keep exploring

Another way into the work.