Skip to content

Pluggedspace Atlas: User Guide & Workflows

This guide provides step-by-step instructions on how to use Pluggedspace Atlas to import datasets, analyze data using natural language or SQL, run automated pipelines, and manage team access.

Team Roles & Permissions

Atlas uses role-based access control to ensure team members have the appropriate permissions:

Role Ideal For What They Can Do
Workspace Owner Account Owners & Team Leads Manage subscriptions, billing, workspace settings, and invite Admins.
Admin Operations & Data Managers Create domains, configure automated data pipelines, monitor ingestion health, and manage datasets.
Analyst Business Analysts & Data Teams Query datasets with SQL or Natural Language, generate reports, configure predictive models, and export results.
Viewer Stakeholders & Executives Search and explore published datasets, view metadata schemas, and receive automated alerts.

Core User Workflows

Workflow 1: Importing and Managing Datasets

  1. Navigate to Datasets: From your Atlas dashboard, click Datasets in the left navigation.
  2. Upload Data: Click + New Dataset and select your file (.csv, .xlsx, or .parquet).
  3. Configure Dataset Info:
  4. Give your dataset a descriptive Name and Domain/Category (e.g., Sales, Marketing, Operations).
  5. Add searchable Tags and a description for your team.
  6. Automated Validation: Atlas scans the file for formatting and quality issues, infers the schema automatically, and assigns it an active version tag (e.g., v1.0).
  7. View Schema & Metadata: Inspect column data types, row counts, and summary statistics on the dataset overview page.

Workflow 2: Querying Data with Natural Language & SQL

Atlas allows both technical and non-technical team members to extract insights immediately.

Option A: Natural Language Querying (NLP)

  1. Open any dataset and click Explore with AI.
  2. Type your question in plain English into the search bar:
  3. "What were our top 5 selling products last month?"
  4. "Find all customer accounts created after March with over $500 in spend."
  5. Atlas translates your query and displays the filtered data table instantly.

Option B: SQL Editor

  1. Click the SQL Studio tab.
  2. Write and run standard SQL queries directly against your datasets: sql SELECT region, SUM(revenue) AS total_revenue FROM sales_data WHERE year = 2026 GROUP BY region ORDER BY total_revenue DESC;
  3. View results in the interactive table or click Export (CSV/JSON).

Workflow 3: Running Predictive Intelligence & Models

  1. Navigate to Intelligence Models from the sidebar.
  2. Select a pre-trained model suitable for your business goal (e.g., Customer Churn Predictor, Revenue Forecaster, Anomaly Detector).
  3. Map Input Data: Choose the target dataset and match the required columns.
  4. Click Run Prediction: Atlas processes the dataset and appends predictive scoring and confidence metrics directly to your results.

Workflow 4: Setting Up Automated Data Pipelines

Automate recurring data transformations so your reports are always up to date.

  1. Go to Pipelines > Create Pipeline.
  2. Select the Source Dataset (e.g., daily sales uploads).
  3. Define the Transformation & Intelligence Rules (e.g., filter out test transactions, apply currency conversion, generate predictive scores).
  4. Set the Trigger Schedule (e.g., Daily at 02:00 UTC or Upon New File Upload).
  5. Click Activate Pipeline. You can monitor run history and status logs from the Pipeline Monitor.

Workflow 5: Exporting & Integrating Insights

  • Manual Download: Click Export on any query or dataset view to download as CSV or JSON.
  • Agent Integration: Share Atlas datasets directly with Pluggedspace Console Agents (e.g., Monica for customer support context or Marketer for audience segmentation).
  • API Access: Use your Atlas API Key to query datasets programmatically from your own applications or BI dashboards.

Best Practices

  • Keep Data Clean: Use standard header names in CSV/Excel files to optimize automatic schema detection.
  • Tag Datasets Thoroughly: Adding relevant domains and tags makes it easy for your whole team to find assets.
  • Use NLP for Quick Checks: Start with Natural Language prompts to quickly explore hypotheses before writing detailed SQL queries.
  • Automate with Pipelines: Instead of manual re-uploads, schedule automated pipelines to keep business metrics fresh.