SynaptaGrid Automation
Build a process visually, connect any system that has an API, and see every run that happens.
Chain steps into a workflow with conditional branching and a human decision where one belongs. Any service with an API description becomes a step you can use. Runs retry, resume and keep their history, and operations can pause or throttle a workflow without a deploy.
Read the Automation documentationBuild-versus-buy boundary
This replaces platform machinery, not product judgment. The division should be clear before the feature catalogue starts.
Infrastructure replacement, not outsourced product design. Review security and deployment →Review portability and continuity →
Define the work
Create the process, publish a version and keep the definition visible to the people who operate it.
Connect the work
See each connection, its protocol and its operational controls before it becomes part of a process.
When it breaks
A three-day approval should not have to happen twice because a downstream API was briefly down.
Run 4,182
Supplier invoice review
trigger · +0s
took 4.2s
approved by M. Okonkwo
502 from the finance API after 3 retries
The approval above took a person and three days. Re-running the workflow from the top would ask for it again. These verbs exist so you do not have to.
Restart the step
The finance API is back. Retry that step in place — the approval stands.
Rewind
The document was read wrong. Go back to that step; everything after it is marked superseded.
Replay
You want the whole run again with the same input, recorded as a replay of this one.
Signal
The run is waiting on something external that has now happened.
A run whose conditions match nothing stops and waits for an operator rather than guessing. These are how you deal with it.
When it matters
Who approved it, that they proved who they were, and that nothing has changed since.
Decision required
Approve settlement · CLM-1043
Your vocabulary, not a generic approve/reject.
Confirm it is you
Re-authentication at the moment of signing, with a second factor where the decision warrants it.
Chosen from codes you define.
For regulated approval it is not enough to know a record was approved. You have to show who approved it, that they proved who they were at that moment, and that nothing has been altered since.
Process
Draw the process, decide what triggers it, and stop for a human decision where a human decision belongs.
6 capabilities
Lay out steps on a canvas and connect them with labeled, conditional links — the diagram is the thing that executes.
Each link can carry a condition; the first matching link wins, and a link with no condition is the default path. A malformed condition refuses to match rather than guessing, so a mistake stops the run instead of silently taking the wrong branch.
Before a version goes live, the workflow is checked for unreachable steps, dead ends, and human decisions that lead nowhere.
Publishing freezes a version, and a run stays on the version it began on — so changing the workflow never rewrites what is already in flight.
A workflow can be triggered by a record changing, by a signed call from another system, on a schedule, or by hand.
A trigger will not arm until its field mappings are complete. That gate is deliberate: a half-configured trigger that fires is worse than one that refuses to.
A step can pause for a human decision, with the task appearing in an inbox and the answer choosing which path the workflow takes.
The decision options and the form are fixed at the moment the task is created, so editing the workflow later never reshapes a task somebody has already opened.
A decision can require the person to re-authenticate and state why, producing a tamper-evident signature record.
This is built for regulated approval, where it is not enough to know that a record was approved — you have to show who approved it, that they proved who they were at that moment, and that the record has not been altered since.
Operations
Every run recorded step by step, with the controls to restart, rewind, throttle or pause without a deploy.
5 capabilities
Follow a run through its steps with the inputs, outputs, errors and the path it chose recorded at each one.
Restart the single step that failed, rewind to an earlier point, or replay the whole run — rather than re-running everything by hand.
A run whose conditions match nothing stops and waits for an operator rather than guessing. That is deliberate, and these are the verbs for dealing with it.
Run state lives in the database rather than in a process, so work resumes after a crash and multiple workers never pick up the same step twice.
Pause, stop or rate-limit a whole connected system or one single action, live, while runs are in flight.
Notifications on failure, completion, timeout, a stalled run, or a failure rate crossing a threshold.
Integration
Any service with an API description becomes usable, and everything that happens here can be pushed back to your systems.
6 capabilities
Register a service, point it at its API description, and choose which of its operations become steps you can use.
The catalog is open-ended by design. What ships pre-built is a starting point, not the boundary — the point is that you are never waiting for someone else to build a connector.
Map a step’s inputs from the triggering event, an earlier step’s output, the record itself, a fixed value or a computed one.
Mappings are compiled when the trigger is armed, so what runs is exactly what was checked — not a fresh interpretation at run time.
A step can read the current state of a record mid-run, so a decision uses what is true now rather than what was true when the run started.
A step can hand work to a service that takes minutes or hours and park until that service calls back.
Subscribe to workflow, step and task events and have them delivered to your own endpoint, queue or stream.
Drop the task inbox, the decision panel and the run timeline into your own interface as ready-made components.
One record, one continuous trace
Data governs the record. Automation runs the work around it. Publish a Data model and its record operations become available to Automation; bind the entity, map the fields and arm the workflow without rebuilding a connector.
Change the model, not the release plan →
Trigger contract
The same record, now armed for work.
Data
Publishing the model makes its typed record operations available in Automation. There is no connector to rebuild.
Automation
Choose the record and the lifecycle moments that matter — Created, Updated or Deleted — then arm the binding.
Automation
A step can read the record fresh just before it acts, instead of relying on an older event snapshot.
Both
Every decision, retry and explicit record update remains visible in the workflow run history.
Procurement, invoices, contracts, onboarding and more — each followed from an empty account to a running process, in the order somebody would do it.
The trial runs the same product described on this page.