AI-powered legal research for lawyers. Legal data APIs for legal tech companies.
For lawyers: chat with your documents, search 100M+ cases.
For legal tech companies: stop building your own RAG pipelines and indexing millions of case laws. We have already indexed 100M+ cases across courts, and expanding to every major country. Plug into our APIs and get AI-powered judgement answers, boolean keyword search, and citation tree graphs with a single API call.
This is the 2nd launch from Vaquill AI - Legal Research & APIs. View more
Vaquill AI - Legal Suit for law firms
Launched this week
Indian legal data infrastructure via API. 20M+ court cases, 23,122 acts, citation graphs, vector embeddings, and legal translation in 11 Indian languages. Built for legal tech companies, AI labs, and anyone building products that touch Indian law.
Free Options
Launch Team / Built With
Hey PH! I'm Priyansh, CTO Vaquill AI.
Some context on why Indian legal data matters right now:
India has the world's largest common law system by volume. 40M+ pending cases, 25 High Courts, 14 Tribunals, 1.7M+ advocates. It's also one of the most underrepresented legal jurisdictions in AI training data.
The two incumbents (SCC Online and Manupatra) have dominated Indian legal data for decades. Neither offers API access, vector embeddings, or citation graphs. They sell subscriptions to lawyers, not data to developers.
In January 2026, Manupatra signed an exclusive partnership with Legora ($5.5B valuation, Swedish legal AI company). SCC Online is partnered with Harvey ($11B valuation). That means if you're building anything that touches Indian law and you're not Harvey or Legora, you have a problem. The data sources are getting locked up.
That's where Vaquill comes in.
I spent 2 years indexing 20M+ cases from every Indian court and tribunal. Built citation graphs across the entire corpus. Embedded everything with Voyage AI for semantic search. Added 23,122 acts and statutes with amendment tracking. Built a legal translation service (Anuvad) covering 11 Indian languages with 22,000+ terms from the government's official legal glossary.
And I made it all available via API.
Some things I'm proud of:
- The citation graph. No other Indian legal database offers machine-readable citation treatment analysis (followed/distinguished/overruled) across 20M+ cases.
- The translation accuracy. Google Translate turns "bail" into "land" in Hindi (जामीन vs जमीन, one letter difference). Anuvad gets it right because legal terminology is not guesswork.
- The MCP server. vaquill-mcp on PyPI gives Claude direct access to Indian case law, acts, and citation networks. If you're building with Claude, you can search Indian law in one tool call.
Some things that are hard:
- Solo founder against incumbents with 30+ year head starts and 100+ person teams.
- Indian legal data is messy. 25 High Courts, each with a different website, different format, different level of digitization. Some tribunals don't even have searchable databases. Cleaning this data was the hardest part of the entire project.
- Monetization is early. Most of my current revenue comes from WhatsApp. Yes, Indian lawyers send me documents on WhatsApp and I translate them and send back the PDF. That's the reality of legal tech in India.
What I'm looking for from the PH community:
1. If you're at a legal AI company (Harvey, Casetext, Legora, or a startup), and you need Indian legal data for your product, let's talk. API access, bulk licensing, or white-label.
2. If you're at an AI lab (Anthropic, OpenAI, Google, Sarvam) and need Indian legal training data, I have 20M+ public domain court judgments with structured metadata and bilingual translation pairs across 11 languages.
3. If you're building background verification, fintech, or compliance products that need Indian court record checks, the API is ready.
4. If you're an Indian lawyer or legal tech founder, try the platform and tell me what's broken.
Thomson Reuters paid $6.5M for TimeBase, an Australian company with 167K legislative items and 11 employees.
I have 78x more data points. Not saying that to boast. Just saying the market values jurisdiction-specific legal data.
Happy to answer any questions. DMs open.