ecommerce_ops · ecommerce · workflow
How Cherrypick built a robust LLM-powered meal generator: lessons from production
Cherrypick's existing meal generator offered no way for customers to specify meal preferences, and the team wanted to add personalization and explain recipe-selection decisions — something difficult to achieve without LLMs.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Customer requests meal plan
A customer initiates meal plan generation by selecting the number of meals they want for the week.
Tools used
Liquid templates
Outcome
The LLM meal generator launched and customers changed their plans 30% less and used their plans in baskets 14% more.
What failed first
A 2023 WhatsApp-based proof of concept for chat-driven grocery shopping was abandoned because the cost per shopping session would have entirely eroded the profit margin.
Results
Volume30% less
Running sincelast month (relative to article publication)
Grounding & classification
Source type: technical build writeup
20 fields verified against source quotes, 1 dropped as unverifiable.
content generationpersonalizationproduct catalogfailure mode describedhuman review describedmetric backednamed customerproduction runtime claimedtools describedworkflow describedecommerceconversion increasecustomer satisfactiontechnical build writeupecommerce opsai draft human approval