The FIR reader
PDF, Word or a scan, in any Indian script. Text is lifted in the browser and only the extracted text is posted to the server, so the scan itself never crosses the network. A corrupt Tamil text layer is detected and the document is sent to optical character recognition instead of being read as garbage.
What it lifts out
FIR and crime number, district, station, occurrence day and date, place of occurrence, type of information, despatch date to court; complainant and accused kept distinct with age, occupation and address; the acts table read as a table; mobile and WhatsApp numbers, bank accounts, UPI handles, IMEI, vehicles, NCRP acknowledgements and UTR references; and each transfer reconciled against the stated total.
AI Case Brief
A structured reading of the report in plain language — what happened, when and where, who complained, who is named, which sections are invoked, how much money moved and where it went. Numbers are attributed to a role, so an action aimed at the accused is never aimed at the victim.
AI Findings
Fourteen to seventeen finding cards, each a specific observation with the evidence behind it and a link through to the record. On a mule case they include the pattern itself: accounts opened in students’ names, rented monthly, sitting in a layer between the victim and the beneficiary.
AI Direction
Eight prioritised steps in the order they should be taken, each naming the record it needs and what it should prove — preserve before the window closes against the accused-facing numbers only; requisition CDR and IPDR separately; serve the bank and obtain the account-opening file; resolve the operator after portability.
Cases board
Kanban across In Progress, Under Trial and Closed, with drag to move, one search across every field, and the photograph of every officer assigned. A case appears only to its creator, its assignees and supervisors.
Officers and roles
Several officers per case at once — lead investigating officer, investigator, analyst, reviewer — which is how a district team actually works.
Evidence explorer
A three-column workspace over a phone extraction: chats read as conversations, calls, messages, contacts, media, web history, wireless networks, application logs, accounts, locations and a combined timeline. A forensic hash is required on every exhibit.
Cross-correlation
Every coordinate the device recorded compared against every tower in the case by true geographic distance, and every party in its chats matched against the call records on the last ten digits. Each match scored, mapped and pinned into the narrative.
Findings, narrative, report
Everything starred anywhere in the case, categorised and ordered, becomes the fourteen-section court report — editable in place, carrying the Section 63 certificate.