PrintBrief: Turning Messy Customer Messages into Print-Ready Orders with Local AI

Описание и применение

PrintBrief автоматически преобразует неструктурированные сообщения клиентов в точные заказы на печать, проверяя и уточняя детали, а затем генерирует точную цену и рабочий лист, сохраняя все данные локально без облачных запросов.

Задача

Print shop owners receive ambiguous customer messages that lack key printing details, leading to costly mistakes and manual clarification.

Результат

PrintBrief automates extraction of print orders from chaotic messages, reduces manual entry, and generates accurate quotes and job slips locally without cloud inference fees.

Рабочий процесс

• 1. Customer sends informal message (e.g., "Bhaiya 25 copies nikal do, A4, kal chahiye.")
• 2. PrintBrief runs a local Gemma 3 4B model via Ollama to extract order fields
• 3. Application validates output with Zod and checks for contradictions or missing info
• 4. Owner reviews extracted fields, answers prompts for missing details, and corrects any errors
• 5. Owner uploads a PDF for page counting (browser‑based pdf‑lib)
• 6. System calculates deterministic quote in integer paise based on shop rates
• 7. Owner confirms order
• 8. Order is saved in IndexedDB with quotation snapshot and downloadable PDF job slip

Доказательства

PrintBrief is a local-first web application designed for neighbourhood and campus print shops. It bridges the gap between chaotic customer WhatsApp/email messages and precise print production specifications—extracting requirements with a local open-weight LLM, counting PDF pages purely in the browser, calculating deterministic quotes in integer paise, and generating official printable job slips.

Источник

Print shop owners face frequent friction when receiving customer orders: Chaotic Messages: Customers send ambiguous text messages like "Bhaiya 10 copy nikal do kal 10 baje" without specifying single vs double-sided, colour vs B&W, paper size, binding, or AM vs PM.

Источник

The open-weight model performs the central task: interpreting informal customer language and converting it into structured order fields.

Источник

Pricing is calculated in code using integer paise. For example, an 11-page document printed double-sided in three separate sets requires 18 physical sheets and 33 printed sides.

Источник

Local inference makes the product practical in three ways. Customer text stays local. With Ollama running on the same computer, the extraction workflow does not need to send customer messages to a third-party AI service.

Источник