What we are building next.
DataCrafter is in active development. This page says what is coming and the month we are aiming at, so you can work out whether the thing you need is close or a long way off.
- Cross-Set paths and dependencies
- The rebuilt browser client
- Data Forms
- Detectors
- Delegations
- Task Scheduler
- Reporting Calendar
- Module version control
- Data adjustment auditing
- Scripting and user functions
- Regulatory reporting
Read these dates as intentions, not commitments
- Nothing on this page has shipped. Every item below is unreleased.
- A date is our current best estimate on the day you read it. Estimates move, and this page moves with them.
- We do not tie a version number to a date. What ships, ships when it is right.
Two pieces of groundwork that most of what follows sits on.
Cross-Set paths and dependencies
Build one processing chain out of several Sets, each on its own lifecycle: sources in one, models in another, reports in a third. A Set names the others it draws on, and the platform tracks what depends on what across the boundary.
The rebuilt browser client
The web client rebuilt in React. Same server, same objects, same results, with an interface meant to feel like the desktop workbench rather than a cut-down version of it. Build work completes at the end of the month.
The four modules listed under On the way on the front page, plus two pieces of control that belong with them.
Data Forms
Hand-entered data and adjustments through a form you author, written back down the same governed path as everything else.
Detectors
Standing checks over a model that raise findings rather than fail a run. Data quality, outliers and rule breaches in one place.
Delegations
Hands a step to an external runtime, Python first, and takes the result back as an ordinary generated model. For the work SQL genuinely cannot express.
Task Scheduler
Fires a chain of work on a calendar or an event, with retries, overlap rules and a notification when something needs you. A stable jobs API and a launch script let an enterprise scheduler drive DataCrafter from outside.
Reporting Calendar
What is due, when, and who has signed it off. Sits above the Scheduler and tracks the obligation rather than the job, so a missed return is visible before the deadline rather than after it.
Module version control
Sets already branch a whole body of work. This adds history to the single object: what changed between two versions, who changed it, and a way back to an earlier one.
Data adjustment auditing
Every manual adjustment recorded against the run it touched, with who made it and when. The adjustment trail becomes part of the audit record instead of a note kept beside it.
Scripting and user functions
Write your own functions and call them wherever a formula or a rule is accepted, for the logic the built-in modules do not cover. Design work has not started, so this one is a direction rather than a plan.
Regulatory reporting
A submission layer over the models you already build, with the regulator-specific part kept to the edges. APRA's prudential returns come first, as validated XML. XBRL arrives with the international regulators that ask for it.
Is the thing you need missing, or further down the list than you would like? Tell us. What people actually ask for changes the order.
Get in touch →