Generation is the easy part. This is what it generates onto.
Genwright does not write a new application stack for every customer. It configures the Knomatic platform — a governed backend that has carried operational workloads for industrial, logistics, maritime and fleet businesses for twelve years.
On this page.
Configuration on a governed platform, not code into an empty repository
An AI coding tool produces source that someone then has to secure, host, test, monitor and keep alive. Genwright produces configuration against a platform where all of that already exists and has already been inspected.
The practical difference shows up on the second Tuesday. When a generated application needs a permission change, a new field, an extra approval step — a code-generation tool regenerates and you re-test the surrounding system. On Genwright the data model, the permission engine and the audit trail are platform services. They do not get regenerated, because they were never generated.
An application generated onto a governed platform inherits its security posture. An application generated into an empty repository inherits nothing.
What every Genwright application gets, without asking
These are not features you enable. They are properties of the platform, present in every application, and they cannot be switched off per app.
Records & relationships
A governed data model with enforced types, constraints and referential integrity. The platform rejects a malformed write; it does not rely on application code remembering to validate.
Roles & row-level security
Access is defined per role and enforced per row. A dispatcher sees their terminal, an auditor sees everything read-only, a contractor sees only their own work.
Audit trail
Who created a record, who changed it, when, from where, and what the previous value was — captured at the platform layer on every record, with no instrumentation step to forget.
Offline-tolerant field capture
Entries made without signal are held locally and synchronized when the link returns, preserving the original timestamp and author rather than the sync time.
Automations & escalation
Rules that fire on record events: open a work order from a failed inspection, alert before a credential expires, escalate an overdue job to the next person up.
Reporting & export
Dashboards, real-time reporting and export to PDF or spreadsheet, scoped by the same roles that govern the records themselves.
Nothing reaches the field until it has been somewhere else first
Every deployment has separated environments. A change is built and reviewed away from live data, inspected at pre-flight, then promoted.
- BUILD
Where the work happens
The Studio operates here. Schema changes, new screens, revised automations — all against representative data rather than the operation's live records.
- REVIEW
Where the operation looks at it
The people who will actually use the workflow try it before it exists in production. Findings go back to build; nothing is promoted on a demo alone.
- PRODUCTION
Where the record lives
Promoted only after pre-flight clears. Backed up, monitored, and carrying a change history that says what moved, when, and who approved it.
The single most common failure of AI-built software is that build and production are the same place. Separating them is not sophistication; it is the minimum bar for a system that operations depend on.
The people who use it every day should not have to be trained on it
Genwright generates the working interface alongside the data model: mobile forms for the field, dense tables for the office, dashboards for whoever is accountable.
Where an operation has its own conventions — a screen that has to look like the paper form it replaces, a portal that has to match a customer's branding, a handheld layout shaped by gloves and sunlight — the frontend is built to that rather than to a template.
Field
Mobile capture designed for one hand, poor light and no signal. Short forms, large targets, photo and signature attachment.
Office
Dense working surfaces: queues, exceptions, bulk actions, filters that match how a dispatcher or coordinator actually searches.
Portals
Scoped external access for customers, carriers, contractors or auditors — the same records, filtered by the same row-level rules.
Genwright rarely replaces everything, and should not try to
Most operations keep their accounting system, their TMS or ERP, their telematics and their payroll. Genwright applications read from and write to those systems over documented interfaces.
Systems of record you keep
Accounting and payroll, ERP, TMS, telematics and ELD, HRIS, document storage. Genwright fills the gaps between them rather than asking you to rip them out.
How the connection is made
Documented interfaces, scheduled or event-driven, with failures surfaced as records rather than swallowed in a log nobody reads.
Why it is scoped early
Integration is usually what determines a project timeline, not the build. It is agreed during scoping so the estimate you get is the estimate that holds.
Platform questions
Is Genwright low-code, or is it code generation?
Neither, exactly. Low-code platforms give you a fixed component set to assemble within. Code generation gives you source you then own and maintain. Genwright generates configuration against a governed platform: the shape of the application is not constrained to a component catalog, but the data model, permissions, audit trail and hosting are platform services rather than generated code.
What happens when we need a change after launch?
The same route as the original build: describe the change, build it in the Studio against the build environment, pass pre-flight, promote to production. Because permissions and audit are platform services rather than generated code, a change to a screen or a rule does not require re-testing the security model underneath it.
Can our own developers work on it?
Yes. Builders work through Anthropic's Claude Code connected to the Knomatic platform, so a technical team can operate it directly. Many operations instead run the first build with a certified builder or channel partner and take it over afterwards.
Where does our data live, and can we get it out?
Operational data lives in your deployment on Knomatic infrastructure and exports in standard formats. Export is a platform capability rather than a support request, because a system you cannot leave is not a system you can safely depend on.
How many applications can we run?
As many as the operation needs. Pricing scales on a platform footprint measured by deployed users or units under management, not per application, so adding a second and third workflow to an existing deployment does not multiply the platform fee.
Bring a workflow and we will tell you honestly whether this fits it.
Not every problem belongs on Genwright. A scoping call is where that gets decided, with someone who has seen the pattern before.