AI Case Studies:
Live Products We Built

Trade law practitioners drowning in hundreds of unstructured legal documents

Legal teams handling anti-dumping investigations, CVD cases, and trade remedy proceedings before the DGTR (Directorate General of Trade Remedies) had to manually search through hundreds of Trade Notices, DGTR guidelines, WTO agreements, and historical case findings — scattered across emails, shared folders, and government portals.

Researching a precedent for an ongoing investigation could take an associate a full day. Tracking multiple live case deadlines simultaneously was error-prone. There was no central intelligence layer to query across all documents at once.

• 60+ Trade Notices with no unified search or context-aware retrieval
• 100+ Anti-Dumping case findings needing manual cross-referencing
• Multiple parallel investigations — deadlines tracked in spreadsheets
• DGFT notifications monitored manually — no alerting or AI query layer

A full-stack RAG AI dashboard — mylegalware.com

We designed and built mylegalware.com — a comprehensive RAG-powered (Retrieval Augmented Generation) AI legal intelligence platform purpose-built for DGTR trade practitioners. Every document collection is queryable via an "Ask anything" AI interface with full document context loaded.

• Law Resource Library — 61 Trade Notices, DGTR Guidelines & WTO Agreements, all AI-queryable
• Findings Database — 102+ Anti-Dumping case findings, CVD & Safeguard cases with contextual AI search
• DGFT Tracker — 42 DGFT notifications monitored and queried via AI
• AI Workbench — draft, analyse, and refine legal documents with AI assistance
• Case Management — live case repository tracking active investigations, deadlines & drafts

A health insurance claim rejection letter most policyholders can't decode

When a health insurance claim is rejected in India, the policyholder usually receives a short letter that cites a clause, an exclusion or a waiting period in insurance language. Whether that rejection holds up depends on the exact wording of their own policy and on IRDAI rules — two things most people have never read.

Even when they decide to challenge it, they have to work out which documents are missing and which letter to send at each stage: a reconsideration request, a formal appeal, a grievance to the insurer's Grievance Redressal Officer, or a complaint to the Insurance Ombudsman.

• Rejection reasons written in policy jargon, with clause references that are hard to check
• The real answer sits inside a long policy document — waiting periods, exclusions, sub-limits, room-rent caps
• Missing hospital or insurer documents that stall a claim
• A multi-step escalation path, each step needing its own formal letter

An AI insurance claim analyzer that reads the documents for the policyholder

We designed, built and operate claimresolveai.com, presented to users as ClaimBhai AI. A policyholder uploads their rejection letter, and the AI extracts the stated reason, explains it in plain language, and maps it to the relevant IRDAI regulations. The product is free for Indian policyholders.

Six tools follow the journey from a rejection to a resolution:

• Rejection Letter Analyzer — extracts the rejection reason and explains it in plain language
• Policy & Exclusion Checker — compares the rejection with the policy wording and shows the clause, waiting period or exclusion involved
• Claim Document Checklist — builds a checklist for the claim type and flags what is missing
• Appeal & Grievance Letter Generator — drafts a reconsideration, appeal, grievance or Ombudsman letter, downloadable as a PDF
• Ask My Policy — AI — answers questions grounded in the user's own uploaded documents
• Claim History — saves every analysis, check, checklist and letter to reopen later

Document AI built for insurance claim analysis

Text is extracted from each uploaded PDF using OCR, including scanned and image-based letters. A retrieval step then pulls matching passages from an indexed library of IRDAI master circulars and regulations, and a large language model produces a structured report from the rejection letter, the policy text and that regulatory context.

• Prompts are written for readers with no insurance background — short sentences, every technical term explained
• The policy checker only quotes wording that actually appears in the uploaded policy, and returns “unclear” instead of guessing
• Its findings are worded as a reading of the text (“the wording appears to support…”), not a legal ruling
• Uploaded PDFs are deleted after text extraction; the extracted text and results are kept in the user's Claim History

From rejection letter to appeal, step by step

1. Upload the insurer's rejection letter, and optionally the policy document. Claim details such as hospital, amount and date are optional.
2. Read the analysis — the stated reason, a plain-language explanation, IRDAI rules that may support the case, a claim strength rating and prioritised next steps.
3. Check the policy to see whether its wording appears to support the rejection, and which clause is involved.
4. Close document gaps with a checklist of what to get from the hospital and from the insurer or TPA.
5. Generate the letter for the current stage, edit it, and download it as a PDF with enclosures listed.
6. Ask follow-up questions about the policy and track everything in Claim History.

For users who only need a letter, a free Guided Letter Builder needs no sign-up: it asks plain-English questions one at a time and fills the letter in live.

What policyholders get

• A plain-language explanation of why the health insurance claim was rejected
• A way to check the insurer's reason against their own policy wording
• The right letter for each escalation stage, ready to edit and send
• A clear list of missing documents before resubmitting
• Free access, with the Guided Letter Builder available without an account
• A library of plain-language guides on claim rejection reasons, IRDAI rights and appeals on the ClaimResolveAI blog

ClaimResolveAI states clearly that it provides guidance, not legal advice.