Work · Manufacturing and distribution

Customers get an answer at any hour, in the language they asked in.

Customers are small-flock poultry farmers, and they ask the same things constantly: what a bag costs, which branch is nearest, what to do about a sick bird. Every one of those questions went to a person, in whichever of three languages the customer happened to write in, at whatever hour they happened to ask.

Client
A poultry-feed manufacturer in Harare
Our part in it
Built by the studio, on a retainer
When
Live since June 2026
The size of the problem

Why an answer at nine at night is worth anything.

80%+Of rural households keep chickens
70%Of a broiler farmer's cost is feed
7.75mDay-old chicks a month, nationally
60%Of national output is small producers

Household poultry keeping from the Food and Agriculture Organization. The feed share of production cost is from a peer-reviewed study of broiler growers in Chegutu district, and a second study puts it at 62%. The monthly chick figure is the Zimbabwe Poultry Association's own, reported in the national press. The small-producer share is the European Union's, from its assessment of the poultry value chain in October 2022. Every customer of this assistant is somebody for whom feed is most of the cost of the thing they are trying to sell.

What we built

An AI assistant on the company's own WhatsApp number. It answers in the language the customer writes in, quotes any of 236 products by name, names the nearest branch from a shared location pin, and accepts a photograph of a sick bird. Staff edit everything it knows themselves, and a correction is live in about a minute.

What it knows lives in a tree, not a pile of documents. Prices, branches, opening hours and products are nodes with values on them, and the tree always wins over anything found in an attached file. That is what makes a correction take a minute: somebody edits a node, it is re-embedded, and the next customer gets the new answer.

No price is hardcoded anywhere in the system. That is not an assertion, it is an audit: a search across the source, the front end, the scripts and the migrations returns nothing that looks like a price, and the evidence pack is kept with the project.

Every morning an automated tester puts realistic farmer questions to the live assistant and a second model grades the replies. The daily report that comes out of it is how a miss gets found by us rather than by a customer.

The loop that matters is the last one. A miss is found by the morning test or by a customer, somebody edits the node that was wrong, and the correction is live in about a minute without a deploy.
Step by step

What happens between a farmer's message and the reply that reaches them.

One message

It is twenty past seven in the evening. A farmer with sixty birds photographs one of them and sends it to a number they got off a feed bag.

19:23
mhoro, ndine huku dziri kurwara. what can i give them?
[a photograph of a hen]
Stage 1

The message arrives on the company's own number

Not a new number, not an app, and nothing to install. It is the WhatsApp number already printed on the bag, and the customer does not have to be told anything has changed.

The signature is verified, the message is acknowledged immediately so WhatsApp does not retry it, and the actual work happens after that. A customer sees the two ticks straight away and the reply when it is ready.

  • WhatsApp Business API
1 of 7
See it running

The delivered system, recorded

The assistant, in the client’s own brand. Twenty-seven seconds. What a farmer sees when they message the number on the truck.
The conversation as a customer sees it, beside the knowledge tree staff edit.
A drawing of the interface, not a capture. The conversation as a customer sees it, beside the knowledge tree staff edit.

What happened once it was running

It runs in production, serving real customers. Every morning an automated tester puts eight realistic farmer questions to it and a second AI grades the replies.

1,371Exchanges in seven days
410Conversations
3.34Exchanges per conversation
236Products it quotes by name
Seven consecutive days of the live assistant, 5 to 11 August 2026

Measured from seven consecutive days of the assistant's own daily usage reports. The product count and the 46 branch locations it can name are counted from the system's own data, which comes from the company rather than from anything they publish, so the entry describes what the assistant covers rather than asserting a branch count on the company's behalf.

Why this page is short

Every figure on this page traces to something we measured.

This page once carried claims that described the work inaccurately. They came down in an audit of our own copy, and each one went back up only when its numbers could be traced to something measured. A case study you cannot check is worth nothing to you, and one that turns out to be wrong costs us more than it ever earned.