Skip to Content

Job Results

After submitting a job, you are redirected to the job detail page at /jobs/<job_id>. This page provides real-time status updates, execution metadata, and full result inspection.

Real-time status updates

Job status updates are delivered via two mechanisms:

  1. Supabase Realtime — A PostgreSQL change subscription pushes updates to the browser as soon as the job_runs row is modified.
  2. Polling fallback — A 3-second polling interval catches fast state transitions that might occur before the Realtime subscription connects.

Subscriptions are automatically cleaned up when the job reaches a terminal state (completed or failed).

Job status badges

StatusMeaning
pendingJob created, waiting for a worker to pick it up.
runningWorker is actively executing the job.
completedExecution finished successfully. Results are available.
failedExecution encountered an error. Error details are shown.

Job details

The job page displays:

  • Backend — The backend name, provider badge, and device type (simulator/QPU)
  • Device identifier — The provider-specific target ID (e.g., Braket ARN), with a copy button
  • Script — Name of the script that was executed
  • Region — AWS/Azure region where execution occurred
  • Timestamps — Created, started, and completed times
  • Wall time — Total execution duration
  • Quantum time — Time spent on the quantum device (when available)
  • Parameters — The full parameters JSON (shots, algorithm-specific config)

Workflow execution dashboard

For jobs that use Temporal workflows, the results page includes a workflow execution dashboard with:

Summary cards

Key metrics from the workflow result are displayed as cards (e.g., final energy, iteration count, convergence status). These are extracted from the _summary field in the job result.

Execution overview

A quick summary showing:

  • Total number of tasks
  • Number of execution levels
  • Maximum parallelism (tasks that ran concurrently)

Gantt chart

A visual timeline showing when each task started and finished, with color coding:

  • Green bars for completed tasks
  • Red bars for failed tasks

Task names and durations are overlaid on the bars. A time axis at the bottom shows elapsed time from start to finish.

Task timeline table

A detailed table grouped by execution level:

ColumnDescription
LevelExecution level (tasks at the same level can run in parallel)
TaskFunction name
StatusCompleted, failed, or pending badge
DurationElapsed time for that task

Failed tasks display their error messages below the table.

VQE optimization timeline

For VQE (Variational Quantum Eigensolver) jobs, an additional timeline table shows each optimization iteration:

ColumnDescription
IterationStep number (first, intermediate, final)
Theta (rad)Parameter value at this iteration
Energy (Ha)Computed energy
WorkflowLink to view the iteration’s Temporal workflow

When there are more than 10 iterations, the table shows only key iterations by default: first, last, biggest energy drops, and any failures. Click Show all iterations to see every step.

Temporal integration

If the job was executed via Temporal, a View in Temporal button links directly to the Temporal UI for the specific workflow run. This provides access to the full execution history, retries, and detailed task logs.

Capsule creation

Completed jobs display a Create Capsule button that packages the job’s script, parameters, and results into a reproducible capsule. If a capsule already exists for the job, a View Capsule button links to it instead.

Inspecting results

JSON viewer

The full result JSON is displayed in a collapsible code block. You can:

  • Copy — Copy the entire JSON to your clipboard
  • Show/Hide — Toggle the JSON viewer
  • Download CSV — For VQE jobs, download iteration data as a CSV file

Braket task ARNs

For jobs that ran on AWS Braket, the Braket task ARNs are listed at the bottom of the results. These can be used to look up the task in the AWS Braket console.

Failed jobs

When a job fails, the page shows:

  • Error message — A summary of what went wrong
  • Error details — The full error output (stack trace, worker logs) in a scrollable code block
Last updated on