← All projects
AI servicesRUNNING
Vectiq Field AI Services
Nine specialised services behind one LLM gateway: retrieval, knowledge graph, prediction, recommendation, embeddings and vision for farm data
Overview
A farm assistant that answers from general agronomy knowledge is a search engine with a friendlier tone. The useful version answers from this farm's own records. Vectiq Field's AI layer is nine services - agent, LLM gateway, RAG, embedding, vector, knowledge graph, predict, recommend and vision - each deployable on its own, with the gateway as the single place where model choice, keys and cost are managed.
The split matters operationally: the vision service needs a GPU and the RAG service does not, so they scale independently, and a farm that only wants disease detection does not deploy the rest.
Key features
LLM gateway: one place for model choice, keys and cost
RAG over the farm's own activity and treatment records
Knowledge graph linking fields, crops, inputs and outcomes
Vision service for crop disease classification
Prediction service for yield and risk
Recommendation service for irrigation and treatment
Embedding and vector services deployable on their own
GPU-bound and CPU-bound services scale separately
Tech stack
Python · aiFastAPI · backendPostgreSQL · databaseRedis · databaseYOLO · aiDocker · devops