FixControl
Company

Governance for AI-powered operations

FixControl is the governance layer for AI-assisted support and engineering. It runs on top of the tools teams already use, so AI can take on operational work while people keep control of what ships.

Why FixControl exists

Operational work is fragmented, and AI made the gaps obvious

Support, engineering, and deployment each live in their own tools, and people carry context between them by hand. Adding AI without a control layer only speeds the fragmentation up. FixControl closes these gaps with governance rather than another silo.

Fragmented tooling

Support desks, issue trackers, chat, CI, and deployment each hold a piece of the truth. No single place shows the whole operation.

Support disconnected from engineering

A customer problem and its fix live in different systems, so the path from a ticket to a verified code change is manual and lossy.

AI without governance

An agent that can act on production systems without policy, approval, or a record of what it did is a liability.

Poor traceability

When something ships, it is often unclear who decided it, on what evidence, and what changed as a result.

Manual handovers

Context is retyped and re-explained at every boundary between teams. That is slow, and detail gets lost along the way.

Slow incident response

Under pressure, teams stitch together tools and threads by hand — exactly when a clear, governed path matters most.

Design principles

The rules FixControl is built on

These principles are design constraints. They decide what the product does and what it refuses to do.

Built on top of existing tooling

FixControl connects to the systems you already run — Slack, Teams, Jira, Freshdesk, GitHub, GitLab, Argo — rather than asking teams to move into a new one.

Human approval for high-risk work

By default, a code change, an external reply, or a deployment waits for a recorded human decision before it happens. Any autonomy is explicit, opt-in, and policy-bounded.

Explainable AI

Every proposal comes with the reasoning and evidence behind it, so the person approving understands what they are approving.

Evidence first

Proposals arrive with the facts that justify them. That evidence is frozen with the decision, exactly as the approver saw it.

Governance by default

Policy, verification, and approval are on by default. There is no fast path around the controls.

Auditability

Every meaningful action records who approved it, under which role, on what evidence, and what followed. The result is an audit-ready trail.

Kubernetes and GitOps friendly

Delivery is pull-request and GitOps native. Changes ship through review and your existing pipelines, never a silent push to production.

Tenant isolation

Each customer's data is isolated at the database boundary and enforced fail-closed, so tenants can never see each other's work.

Enterprise security

Signed webhooks, secret-safe logging, scoped admin access, and encrypted credentials come standard.

The story behind FixControl

One pattern kept repeating: lost context, unclear decisions

FixControl came from watching the same operational failures repeat across software projects and organizations: in support, in engineering, in deployment, and in how those workflows fit together.

The pattern was consistent. Capable teams were slowed by fragmented tools and manual handovers, and more recently by AI bolted on in ways that removed control rather than adding it. FixControl is built around one conclusion: make AI useful inside enterprise workflows while keeping humans in control.

FixControl is built by Van Uden Investments B.V. in Hoorn, the Netherlands.

How the idea took shape
  1. Observation
    Fragmented tools, lost context

    Very different teams hit the same problems: fragmented workflows, context lost at every handover, and no clear line from a problem to a verified fix.

  2. The AI shift
    Agents could act, ungoverned

    As AI entered these workflows, the gap became concrete. Agents could act, but nothing verified, approved, or recorded what they did.

  3. The goal
    Useful AI, with humans in control

    The aim became a governance layer: let AI propose and prepare the work, but keep policy, verification, and human approval in the path of anything consequential.

  4. FixControl
    Governance became the product

    That layer became FixControl. AI drafts the work, policy and verification check it, and a person approves it. Every decision is recorded and auditable.

Where FixControl is going

One governed layer across the whole operation

The direction is a single governed layer across support, engineering, and deployment. AI takes on more of the work; accountable people keep the decisions.

  1. 01

    Connect existing systems

    Bring support, engineering, and deployment tools into one governed flow, so work moves between them without manual re-entry.

  2. 02

    Improve operational context

    Carry the full context of a problem — evidence, history, and decisions — alongside the work, so nothing is lost at a boundary.

  3. 03

    Govern AI execution

    Keep policy, verification, and approval in the path of every AI action that touches a production system.

  4. 04

    Speed up incident response

    Give teams a single path to follow under pressure, with the same checks in place, so responding faster does not mean cutting corners.

  5. 05

    Maintain human oversight

    Keep a person accountable for every consequential decision, with the evidence they need to make it well.

AI should increase confidence, not reduce control.

Every design decision in FixControl comes back to this. AI in operations is worth adopting when the people responsible can see what it does and stay in control of it.

Company
Van Uden Investments B.V.
Hoorn, the Netherlands

FixControl is a trade name of Van Uden Investments B.V. The company builds one thing: a governance layer that keeps people in control of AI-assisted operations.