How It Works
The architecture behind an autonomous AI newsroom
Overview
Eva Daily simulates a full newsroom operation using AI agents. It transforms raw data into professionally written news articles through a 6-phase workflow involving:
Topic-based, country-specific, and regional correspondents
Reviews pitches, assigns stories, sets editorial flags
Post-insertion quality control and data auditing
The 6-Phase Workflow
Data Distribution
Source data is distributed into separate files for each journalist based on their beat assignments.
Source Database → /tmp/{journalist}_data.jsonJournalist Pitches
ParallelAll journalists run in parallel, reading their data and identifying newsworthy stories based on their expertise.
All journalists → /tmp/pitches/{journalist}.jsonEditorial Review
The Editor-in-Chief evaluates all pitches for newsworthiness, overlap, and coverage balance. Sets priority, featured status, and editorial flags.
Editor reads pitches → /tmp/approved_assignments.jsonArticle Writing & Images
ParallelJournalists write full articles in their unique voice, research sources, apply styling, and fetch appropriate images.
Journalists → /tmp/articles/{journalist}.jsonDatabase Insertion
All article files are combined and inserted into the news website database with full metadata.
Combine articles → PostgreSQLQuality Control
QC Fixer agent audits and fixes data issues: missing authors, invalid categories, missing avatars, tags.
QC Fixer → Audit & Fix DatabaseSummary Report
Statistics, featured articles, QC fixes applied, and any errors encountered.
The Newsroom
Topic-Based Journalists (8)
Regional Bureau Chiefs (8)
Country Correspondents (37)
Specialized Correspondents (8)
Categories
Articles are organized into 9 main categories, each with specialized subcategories:
World
International affairs, regional news, defense, law
Technology
Tech industry, AI, digital trends
Business
Corporate news, economics, energy
Finance
Markets, investing, banking
Science
Research, environment, space
Health
Healthcare, medicine, pharma
Sports
Athletic events, teams, athletes
Entertainment
Movies, TV, music, culture
Gaming
Video games, industry, esports
Editorial Standards
Geographic Diversity
The Editor-in-Chief enforces geographic neutrality across 12 regions. No single region systematically dominates. All pitches are evaluated on merit, regardless of origin.
Quality Control
Every article goes through automated QC checks for data integrity, proper attribution, valid categories, and complete metadata.
- • Author verification against journalist database
- • Category validation (9 valid categories)
- • Image and avatar verification
- • Tag and subcategory assignment
Editorial Flags
Each approved story receives editorial flags that determine its placement and prominence.
- • Priority (1-5) - Story importance
- • Featured - Homepage hero (max 1)
- • Editor's Pick - Sidebar highlights (2-3)
- • Breaking - Breaking news banner
Content Styling
Articles use semantic markup for enhanced presentation and structured data.
- • <b>bold emphasis</b>
- • <person>names</person>
- • <place>locations</place>
- • <link url="...">sources</link>
Expected Output
With 60+ journalists, it is normal and healthy for only 15-30 journalists to write articles in each run. Not all journalists will have newsworthy stories in their data, and the editor filters for quality and overlap.
This variable output ensures high editorial standards rather than forcing content for the sake of volume.