Contact person: Business Client

Phone:Show

Email:Show

Location: Calgary, Canada

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"I Hi! Need you to extract customer info-name, email, address, phone, site-from PDFs, Word files, images via OCR. List sample descriptions like drywall compound, popcorn ceilings, stairway lino, garage stucco; count them. Fill chain-of-custody form, email to right lab. Generate zero invoices with details. Use Node.js-no Python-so I can tweak it myself. Bid?need a small workflow that lifts key data from incoming PDFs, recognises handwritten figures on attached images, then turns all of that into ready-to-send Xero invoices and confirmation emails.

What has to happen
• Parse the email’s PDF attachment and pull out contact details, invoice numbers, dates, line-item descriptions, quantities and prices.
• Run OCR on the accompanying handwritten images so any amounts or notes make it into the same data set. Tesseract, AWS Textract or a similar engine is fine as long as it reads handwriting accurately.
• Use the Xero API to create a draft invoice with the extracted data, apply the correct contact record or create one when it doesn’t yet exist, and attach the source documents for audit trail purposes.
• Auto-populate a simple web form with the same information (our internal log) and trigger an email to the client with the Xero PDF attached.

Acceptance check
An email that arrives with a PDF and a handwritten image should, within a minute, result in:
1. A draft invoice visible in our Xero account containing the correct contact details, invoice details and product or service descriptions.
2. The internal form saved with matching data.
3. A confirmation email sent to the client (copied to us) with the generated Xero invoice attached.

Code can be Python, Node or a low-code platform—whichever you’re fastest with—so long as it’s well commented and I can adjust field mappings later." (client-provided description)