Where is our personal data?
Across databases, exports, spreadsheets and vendor systems. Not where you think it is — where it actually is.

A data discovery, classification and erasure engine that runs inside your network — and closes every deletion request with a certificate naming what was erased, what was kept, and under which statute.
Lekha is in build. It runs end to end — every screen in the walkthrough is captured from a live deployment — but on a synthetic estate: no client system has been connected to it yet, and no external auditor has reviewed an evidence pack. We are choosing the first installs now, and would rather tell you that here than in the third meeting.
Across databases, exports, spreadsheets and vendor systems. Not where you think it is — where it actually is.
Some data must be deleted on request. Other data must be kept for years because another law says so. Both are true at once.
Not a promise in a policy document. Evidence a third party will accept, produced by the system itself.
Generated by your systems on a schedule, with the evidence behind every finding — and no raw personal data stored.
Aadhaar, PAN and GSTIN confirmed by checksum, not by a guess at the column name.
Each clause naming its statute, its period and what happens at the end. Your counsel can review every line.
For example the Prevention of Money-Laundering Rules, 2005 — not just “five years”.
Certificates, a chained record of every action, and one archive an examiner reads. Not a report written afterwards.
Each export is logged: your name, the moment, and the exact scope of what left the appliance.
Four of these nine open into a diagram further down. The locator beside each one says which station you are looking at.
The handover Steps one to three and steps four to seven cannot be the same person. The rule sits in the engine, not in a policy document.
One of them wins, and the certificate says which.
Each entry carries the previous entry’s fingerprint. Alter any one of them and see what happens.
The DPDPA readiness assessment stands on its own, and produces the data map a deployment would start from.