Two products.
One thesis.
Conversational commerce is the next channel. We build the engine and the merchant-facing assistant on top of it. Iris is what shows up in Messenger. Nous is what runs underneath.
Conversational commerce for f-commerce merchants. Messenger-first, Bangla and English.
300,000+ Bangladeshi merchants run their entire business through Facebook Messenger DMs — manually. Iris is the AI assistant that takes orders, answers questions, dispatches couriers, and chases payments. Bangla-native, merchant-priced, built on Nous.
- —Bangla + English, code-switched. Handles the way merchants and customers actually talk — Banglish, voice notes, broken sentences.
- —Order-to-courier in one chat. Captures address, weight, product details and dispatches to Pathao / Steadfast / RedX without leaving Messenger.
- —Payment chase + collection. Sends bKash / Nagad payment links, follows up on unpaid orders, flags COD risk.
The channel-agnostic conversational commerce engine. Iris is built on it.
Nous is the conversational commerce engine that powers Iris and (soon) other channel surfaces — WhatsApp, Instagram DMs, web chat, voice. Spring Boot + pgvector + Gemini under the hood. LLM-first architecture: every interaction goes through reasoning, not regex.
- —Channel adapters, one core. Same conversation engine across Messenger, WhatsApp, Instagram, web. Add a channel without rebuilding the brain.
- —LLM-first orchestration. No hardcoded flows. Intent + context + tools, every turn. Cheap models for routine turns, big models when judgment is needed.
- —Merchant-private vector memory. Each merchant gets isolated retrieval — products, FAQs, past customer history. No cross-tenant leakage.
Want to build on Nous,
or sell with Iris?
Iris is in private beta with BD merchants. Nous is the engine — open to platform partners by conversation.