Grounded garden consultant for a regional grower
A garden consultant that retrieves symptom guidance, checks recommendations against the grower's assortment and routes uncertain questions to a specialist.
What changed. Catalog checks and specialist handoff are the release requirements for seasonal product questions; unsupported products or doses must fail review.
Stack. symptom retrieval · grounded generation · dosage guard · messenger bot
Context
A regional grower and garden supplier sells crop-care products, seeds, plants, and tools to a seasonal audience. The buyer does not arrive asking for a fungicide SKU. The buyer arrives with dying cucumbers and a phone full of generic search results. A human agronomist can assess the problem, but availability varies during the spring peak. A generic answer may not reflect the supplier’s catalog or the customer’s conditions. Invented doses on products children eat are not a copy problem. They can harm people, crops and the supplier’s reputation.
Task
Put a consultant on the grower’s own assortment: use a symptom description to retrieve relevant guidance, say what to do this week in the customer’s climate band, and name a product that is actually on the shelf. Unknowns escalate. The bot is a sales instrument bound to inventory, not a public encyclopedia. Voice: a careful neighbour, not a lecture in Latin.
What we built
A reusable circuit with a knowledge base of diseases, pests, practice, varieties, and a calendar — each record tied to the client’s SKUs. The user describes the problem in their own words; retrieval is semantic, not a keyword lottery. Generation is allowed only from retrieved records. An output guard checks named products, doses and intervals against the base. Catalog presence alone does not establish that a treatment is appropriate; current label restrictions and specialist review remain necessary. Failures retry or refuse and pass to a specialist. If the model is down, the bot collects a contact instead of going mute.
The same kernel serves more than one instance. Channels are Telegram and WhatsApp. Dialogues and leads stay with the grower. The calendar is keyed to climate band and crop, not to a named province. Assortment and timing are meant to be refreshed before each season so the bot does not recommend last year’s pack.
What changed
The workflow separates routine catalog questions from specialist review. Availability, correct advice and safe dosing are not guaranteed by the architecture; unsupported or uncertain recommendations require handoff. What people ask — crop, symptom, season — becomes a feed for assortment and content, not a lost search elsewhere.
What we would do differently
Ship one crop and one climate band first, with a hard empty-set refusal, then widen the calendar. The temptation is to sound complete on day one. A broad scope with a thin base creates opportunities for unsupported recommendations. Completeness is a later season, not a launch claim.
Related service: AI consultant and chatbots