Features · Compliance Co-pilot
Where the question comes from. How it's answered. What you can verify after.
Layer 1
The default channel. Each seat is a phone number; numbers are added to your WhatsApp Business account on onboarding. Messages route through Meta's BSP layer. Read receipts, typing indicators, image attachments (for notices), and voice notes (transcribed) all supported.
For teams in regions or contexts where WhatsApp isn't the right channel — finance teams in multinationals often prefer Telegram or a web widget. Identical capabilities, identical audit trail.
A language classifier runs first; the retrieval still happens over the English corpus, but answer generation flips to the requester's register. Tamil / Marathi / Gujarati are Enterprise add-ons; ask sales for the rollout calendar.
Photograph a GST notice; the copilot extracts the section, demand period, demand amount, due date, and surfaces a draft response. Image OCR is opt-in per tenant.
Layer 2
Every question is routed to a regulator-specific retriever (CBIC, RBI, SEBI, MCA, …). Mis-routing is the leading cause of bad answers; the classifier is monitored.
Each regulator's circulars + notifications + master directions are re-indexed weekly. Embeddings + keyword + section-number hybrid retrieval. Stale answers are caught at the citation-validity check.
No answer escapes without at least one citation that links back to a paragraph in the corpus. If retrieval is empty, the answer becomes "I don't know — escalate."
Every inference is captured as a span tree: classifier → retriever → LLM call → validator. Admins replay traces in the Langfuse UI; flag bad ones for next-week's evals.
Layered input classifier + output validator. Caught attempts are logged with classification only — never with full content beyond a hash, per DPDP.
"I don't know" routes to a human escalation queue: your own internal CA panel by default, our partnered CA firm as a backup (Enterprise add-on).
Layer 3
The bottom of every answer carries a Sources block — circular numbers, dates, paragraph IDs. Tap a citation to receive a verbatim quote with the link to the source PDF.
Growth + Enterprise admins get a Langfuse-backed trace UI. See what was retrieved, what was generated, where the citation came from. Tag bad traces; they enter our weekly eval set.
Append-only conversation history per seat. Exportable on demand. Retention is configurable per tier; 30d default on Starter, 90d on Growth, custom on Enterprise.
Every Monday, admins get a digest: question volume, regulators most asked about, top flagged answers, suggested re-training subjects. Helps compliance leads spot the gaps in their team's knowledge.
Comparison
| Capability | Google search | ChatGPT | Sahayak Co-pilot |
|---|---|---|---|
| Cites the source regulation | Sometimes (stale) | Hallucinates | Always |
| India-specific corpus | Mixed | Generic | Yes, re-indexed weekly |
| Lives in WhatsApp | No | No | Native |
| Audit-logged conversations | No | No | Yes, DPDP-compliant |
| Prompt-injection defence | N/A | Limited | Two-layer |
| Admin review of bad answers | No | No | Langfuse trace tagging |
| Hindi / Hinglish | Limited | Limited | Native |
| "I don't know" honesty | Pretends to know | Confidently wrong | Escalates to human |
We'll answer them live on WhatsApp, with citations. If we're wrong, we explain why.