Person using WhatsApp on a smartphone
Tshomi AI

The agent, on a phone,
in every language.

Tshomi speaks first, in the participant’s language. Every message calls a tool. Every tool leaves an audit-trail row. 98% of sessions close without a human.

11
Languages
All 9 SA official + English + Afrikaans
13,164
Tool calls
Last 30 days · live instance
98%
Sessions closed
Without a human · Aug 2026
0
PII in logs
Redacted before any model call
How it thinks

Not a chatbot.
A tool-calling agent.

Tshomi doesn’t just reply. Every turn either calls a real tool (book a slot, verify a document, record a check-in, deliver a payslip) or hands off to a supervisor with the context to act. Nothing it does is invisible.

“Every message calls a tool. Every tool leaves a row.”
Design principle
Tshomi
online · Language switcher
isiXhosa · isiZulu · Sesotho · English
17:54
5G
Two surfaces

WhatsApp for participants.
In-app for supervisors.

Same data, same audit trail. Different surface for each audience.

Person using WhatsApp on a smartphone
For participants
WhatsApp Business

Participants never install anything. Tshomi lives in the WhatsApp thread they already trust — same phone number they use for family and friends.

Live dashboard view
For supervisors
PWA / Capacitor app

Supervisors get a full operations console — same data, same audit trail, richer view. Runs as a PWA in the browser or a native app on their device.

Tshomi acted

Every message calls a tool.

Every tool call is a row in the audit log — exportable JSON, filterable by programme, cohort, participant or period.

13,164
Tool calls
Last 30 days
53%
Auto-routed
Tickets
98%
Closed by Tshomi
Aug 2026 sessions
0
PII in logs
Redacted at ingress
Live tool-call log · sample day
Live
05:47booked_biometric_slotNomsa · Site 042 · Portal 3
05:49chased_documentbank_letter · Nomsa
05:52verified_documentbank_letter.pdf verified
05:55biometric_matchconf 98.4%
06:00recorded_checkinNomsa · 06:00
06:12created_ticketINC-2411 · water_supply
06:12routed_incidentSipho · Facilities · SLA 45m
06:34delivered_payslipwk34 · R2,537.57
Capabilities

What Tshomi does, without a human.

Team collaborating on laptops
In-language
Greets first, in mother tongue

Every participant is met in the language they set — from the first message. The language switches; the tool call and receipt don't.

Precise face recognition abstract
Biometric
Face + voice on sensitive actions

Check-ins, payouts and consent-gated flows require a biometric match. Templates encrypted at rest, revocable on request.

ID document scan
Documents
Chases, receives, verifies

Bank letters, ID copies, contracts — Tshomi asks for them, receives them in WhatsApp, and verifies them before writing the audit row.

Languages

All nine SA official.
Plus English and Afrikaans.

Participants are greeted in their mother tongue on the first message. Same tool call, same audit-trail row — the language switches, the receipt doesn’t.

EnglishisiZuluisiXhosaSesothoSetswanaSepediXitsongaTshivendaSiSwatiAfrikaansisiNdebele
11
Languages on WhatsApp
1
Audit row per message, any language
EnglishisiZuluisiXhosaSesothoSetswanaSepediXitsongaTshivendaSiSwatiAfrikaansisiNdebele
Under the hood

Model stack.
Guardrails.

Person using a smartphone
Model stack

Haiku for triage. Sonnet for narrative.

Fast, cheap model handles routing, classification and structured extraction. Larger model runs for participant-facing narrative in-language. Guardrails at both layers.

Claude Haiku 4.5Claude Sonnet 4.6
Dark screen with code, secure environment
Guardrails

Randomised chat data. Zero PII in logs.

  • PII redacted before any model call
  • Structured outputs validated against JSON schema
  • Tools gated by programme-scoped RLS
  • All prompts version-controlled and diffable
See it running

Book a live demo.

30 minutes. We walk your team through a live thread, show the tool-call log in real time, and generate a compliance pack from your own sample data.