Supreme Court Judgments
75 years of case law, opened to everyone.
The problem
Seventy-five years of Supreme Court judgments sat locked in the ecourts portal — scattered PDFs across many languages, with no clean, queryable, bulk-accessible corpus for researchers or AI teams to build on.
Product vision
An open, machine-readable record of every Supreme Court judgment — free to access, simple to query, and ready for legal research and AI.
Architecture
Judgments are collected from the ecourts portal, normalized, and published to a public AWS S3 bucket (indian-supreme-court-judgments, ap-south-1) as language-tagged archives. Metadata is emitted as raw JSON and columnar Parquet, queryable directly with AWS Athena — no AWS account required. The corpus is refreshed bi-monthly under a CC-BY-4.0 license.
Engineering challenges
- Extracting clean text from inconsistent, multi-language court PDFs.
- Normalizing 75 years of documents into one consistent, queryable schema.
- Publishing at bulk scale with zero-friction, credential-free access.
Performance & optimization
- Columnar Parquet metadata makes Athena queries fast and cheap.
- Language-tagged archives keep bulk downloads targeted.
- Bi-monthly incremental refresh keeps the corpus current without full re-publishes.
The outcome
A free, AWS-hosted corpus of Indian Supreme Court judgments (1950–2025) that anyone can query or train on — listed on the AWS Open Data Registry and maintained by Dattam.
Tech stack
Have a product to build?
Tell us what you're building. We'll show you how we'd architect, design, and ship it, and what it's actually like to work with us. First reply within one business day.
LAT 17.40 · LON 78.48 · contact@dattam.in