About Jobs
In Opaque, a job is a secure, reviewable task that runs on encrypted data within a workspace. Jobs allow you to query data, perform transformations, and generate insights—without ever exposing sensitive information. All jobs are executed inside trusted execution environments (TEEs), ensuring privacy and compliance at every stage.
Opaque supports three types of jobs:
- Notebook jobs
- No-code jobs
While the underlying lifecycle is the same, the way you create and configure a job depends on the job type.
Notebook jobs
Notebook jobs are designed for users who are comfortable writing queries in Python using PySpark or PySpark SQL (a SQL-like query language that runs inside Python using Spark). They provide full control over data access, transformation logic, and output.
With notebook jobs, you can:
- Write and test your own queries in a built-in editor.
- Reference connected datasets and use synthetic test data.
- Define input variables for dynamic execution.
- Review and approve jobs.
- Run complex analyses that go beyond the limits of predefined templates.
Notebook jobs are ideal for data scientists, analysts, or technical users who need flexibility and direct access to the data schema.
No-code jobs
No-code jobs are built from prewritten templates. Instead of writing code, you use a form-based interface to configure inputs like datasets and matching columns.
With no-code jobs, you can:
- Choose a template that fits your use case.
- Input required fields through a guided form.
- Review and approve jobs like any other.
- Run secure analysis without touching code.
No-code jobs are best for business users or collaborators who want to securely analyze sensitive data using predefined workflows.
Which job type should I use?
Use the following table to decide which job type fits your goals:
| Feature | Notebook Job | No-code Job |
|---|---|---|
| Requires coding | Yes (Python or PySpark SQL) | No |
| Workflow customization | Fully customizable | Fixed to template logic |
| Use cases | Advanced queries, ML, transformations | Overlap analysis, filtering, lookups |
| Setup complexity | Higher | Lower |
| Ideal for | Analysts, data scientists | Business users, privacy/compliance leads |
If you’re building something new or need full flexibility, start with a notebook job.
If you're running a common workflow—like measuring data overlap or filtering users—use a no-code job.
Next steps
- Job workflow: Understand the four main stages in the Opaque job workflow.
- Notebook jobs: Learn how to write and test custom queries using the query editor.
- No-code jobs: Use prebuilt templates to run analyses without writing code.