Our Process
How a corruption case goes from discovery to the public archive
Every case on Jawafdehi passes through a structured pipeline — from raw government documents and media sources through research, AI-assisted drafting, human review, and permanent publication.
The Case Pipeline
Discovery
Our data scraping team archives government documents, CIAA filings, court records, and CIB reports into digital text. The outreach team simultaneously works with investigative journalists, corruption watchdogs, and media organisations to surface cases not yet in the public record.
Research
Our research team investigates what constitutes corruption in each case, the role it plays in Nepali governance, and the effectiveness of existing anti-corruption frameworks. Every case is contextualised before it is written up.
Compilation
The compilation team structures each case: key allegations, timeline of events, entities involved, and all source documents. Cases are written in plain language so citizens — not just lawyers — can understand them.
AI-Assisted Drafting
AI helps our team process large volumes of legal documents faster — summarising filings, extracting key facts, and flagging inconsistencies. Every AI-generated output is reviewed and approved by a human volunteer before publication.
Verification & Publication
All cases are cross-referenced against at least two independent sources. Once approved, the case is published to the archive — permanently. Records are never altered without a visible audit trail and never deleted.
Ongoing Tracking
Published cases are actively monitored for developments: new court orders, verdicts, appeals, and entity responses. Updates are added to the existing case record so the full history is always visible.
Where Our Data Comes From
We draw from official government sources and independent civil society — always citing the origin of every claim.