Skip to content
Artificial IntelligenceR&DConversational assistant

Conversational assistant — Telecom

Retrieval-grounded multilingual assistant in Gujarati, Hindi and English. Delivered for a telecom client in Motera, Ahmedabad.

Location
Motera, Ahmedabad
Year
2023
Duration
6 weeks
Team
5 engineers
Interactive 3D · drag to rotate
86%
queries resolved
5.4
languages
2.9 s
response time

The challenge

Existing vendors had delivered pieces of a conversational assistant system, but nobody owned the integration and the pilot had stalled.

What we built

We owned the whole stack — hardware, firmware, models and dashboards — using LLM agents, Vector search, FastAPI so there was one accountable team.

Delivery timeline

  1. Phase 1

    Discovery sprint and site survey

  2. Phase 2

    Prototype built and tested in the Motera lab

  3. Phase 3

    Pilot on site with before/after measurement

  4. Phase 4

    Scale-out, training and handover

Highlights

  • Designed, built and supported from Motera, Ahmedabad
  • Safety case reviewed with the client's EHS team
  • Documentation, drawings and source handed over at close
  • Operator training delivered in Gujarati, Hindi and English

Have a problem that needs a machine built for it?

Tell us the constraint — line speed, accuracy, environment, deadline. An engineer will read it and reply with a straight answer, including when the answer is that you don't need us.

We reply to every enquiry within one business day.