Telecommunications01
Customer support at scale
Omnichannel customer service for a major telecom operator — plan selection, billing and account journeys across millions of subscribers.
Enterprise AI Chatbot
Customer service that resolves itself.
Chatbot is enterprise-grade generative AI that understands intent across languages, resolves issues end-to-end, and escalates to a human only when it genuinely should.
Measured · 4 weeks post-deployment
- 90%
- First-contact resolution
- +70
- NPS point swing
- 63%
- Resolved autonomously
- 40%
- Lower agent cost
up from 60% before deployment
from −39 to the +30s
end-to-end, no agent involved
while absorbing 19% more volume
Conversation designer
Design the conversation. Deploy it everywhere.
Build resolution flows visually — pickups, tracking, fares and human handoff — then ship them to every channel from one place.
Rides Powered by Chatbot
Where can I take you today?Airport — terminal 2Pickup in 3 min24 min trip · $18.50Arman is on the way — white Toyota, B 1234 XYZ.
The platform
Built to resolve, not deflect.
Sophisticated language understanding, wired into your systems, with human escalation when it matters — the combination that moves the metrics.
Intent-aware understanding
Detects intent, extracts entities and holds multi-turn context — so it grasps what a customer actually needs, not just the keywords they typed.
Multilingual & dialect
Auto-detects language and handles regional dialects and transliteration like Hinglish — customers never have to adapt their natural style.
Intelligent escalation
Confidence scoring routes complex or sensitive cases to humans, with full conversation context handed off so customers never repeat themselves.
Omnichannel by default
One assistant across web, app, WhatsApp, Messenger, IVR and government portals — with unified history across every touchpoint.
Enterprise integration
API-first connections to billing, CRM and provisioning systems enable real-time data and transactions, with PII masking and encryption throughout.
Continuous learning
Human-in-the-loop feedback turns agent corrections into knowledge-base and model improvements, week over week.
Case study 01 · Enterprise AI chatbotMajor Indonesian telecom operator (under NDA)
Handling 300K+ monthly conversations at 90% FCR.
A market-leading prepaid operator was drowning in volume — NPS as low as −39, ticket backlogs peaking at 178 a week. Within four weeks of go-live, Chatbot reversed it.
First contact resolution
90%
Up from 60% before deployment.
Monthly conversations
300K+
Across two brands, every channel, at production scale.
Agent handover rate
20%
Down from 32% — one-third of escalations removed.
Automated journeys
80+
From package selection to refunds, across 100+ API integrations.
WhatsAppWebAppMultilingual NLPRAG-based interactionsSeamless human handoff
Case study 01 · continued
Sentiment swung 62 points after deployment.
NPS moved from −32 to +30 post-deployment.
Consistent weekly NPS in the +25 to +35 band — a full sentiment reversal, sustained across the first six months of production traffic.
WK 01WK 06WK 12WK 18WK 24
Enterprise AI Chatbot
Selected deployments · most under NDADeployed across industries.
Automotive marketplace02
Buyers, sellers, listings
Conversational support for a high-volume used-car marketplace — inspection and booking workflows, lead qualification and transaction help.
Deep-tech hardware03
Technical product support
Pre- and post-sales support for a precision-instruments maker — spec lookups, B2B inquiries and high-accuracy deflection.
Founders
Three builders. One thesis.
Backed by experience at
- Samsung Research
- NUS
- National Taiwan University
- Indian Institute of Science
- University of South Carolina
- IIIT Delhi
APhoto
Co-founder · CEOAryaman
From AI & security research at NUS and NTU to shipping production AI systems at Samsung.
HPPhoto
Co-founder · CTOHarsh Pandey
Across NUS, IISc and USC, built foundations in ML; now leads system architecture end-to-end.
SAPhoto
Co-founder · CPOSubhanshu Arya
From ASR research at IIIT Delhi to production voice workflows and live client deployments.
Get started
Bring resolution-grade AI to your customer service.
Tell us about your support volume and stack. We'll show you a flow built for your highest-volume use case — and the metrics it would move.
- Live on high-volume use cases in weeks
- Resolves end-to-end — not just deflects
- Escalates with full context to your agents
Request a demo
We'll tailor the walkthrough to your use case.
FAQ
Questions, answered
What enterprise teams ask before deploying Chatbot.