Capture the fields that matter
Extract the manufacturer lot number, manufacture date and expiry date directly into the fields warehouse teams need.
WMS Label Recognition
Turn label photos into warehouse-ready fields.
Extract manufacturer lot numbers, manufacture dates and expiry dates, with a standard API connecting recognition, human review and WMS entry.
- recognition accuracy
- 100%
- logistics users
- 28
- calls per day
- ~520
A better way to work
BEFORE
Warehouse staff manually locate and type label details while dealing with inconsistent layouts, date formats and similar-looking characters.
WITH THE PRODUCT
A photo produces structured fields for review. Staff confirm or correct the result, and the business application completes WMS entry.
Core capabilities
Keep review in the workflow
Structured results remain editable, with missing information left empty. Staff retain control of the final confirmation.
Fit existing warehouse operations
A PHP JSON API, timeout handling and feedback interface let the calling application connect capture, display, review and entry.
How it works
- 01
Photograph the label
- 02
Extract lot and dates
- 03
Review or correct fields
- 04
Confirm and enter into WMS
Key design decisions
Design the output around three fields
Warehouse entry needs a manufacturer lot number, manufacture date and expiry date. I shaped the service around those structured fields so the business application can display and enter them directly, keeping review focused on the information staff need.
Leave the final check with the user
Recognition is followed by review and correction, with absent label information left empty. Staff retain final confirmation, turning character-by-character typing into a focused review of the extracted fields.
Connect through the existing workflow
A PHP JSON API provides recognition while the existing business application handles capture, display, confirmation and WMS entry. Timeout handling and feedback interfaces let the caller manage exceptions and support manual intervention.
PRODUCT / ENGINEERING
What I delivered
I designed the workflow, compared vision models with OCR-based approaches, and built the demonstration and PHP recognition API. I defined request formats, structured outputs, timeouts and feedback interfaces for integration into the business application.
TRY THE WORKFLOW
INTERACTIVE WALKTHROUGHRecognize. Review. Confirm.
This walkthrough uses sample data and preset results, without connecting to company systems or live AI.
INPUT · SAMPLE LABEL
LAB / SAMPLE01
SAMPLE REAGENTREF. DEMO-A01 · 10 mLLOTLT260901
MFG2026.09.01
EXP2027.09.01
FOR DEMONSTRATION ONLYMake every field reviewable
Select a label, inspect its fields and explore the review, correction and confirmation flow.
Human review complete
- Lot
- Manufactured
- Expires
Walkthrough complete. Choose another label to explore a different scenario.
Does your team face a similar challenge?
Start with one workflow. Let’s explore how AI could fit your operations.
Keep exploring
All workEnterprise fulfillment assistant0→1 · Solo build
LD Assist
An enterprise fulfillment agent powered by a locally deployed large language model. Connect orders, purchasing, warehouse records and follow-ups so support teams can check progress, confirm batches and prepare reply drafts.
- support users
- 30
- queries per day
- ~300
跟一下
Personal follow-up app0→1 · Solo build
跟一下 (Genyixia)
A lightweight iOS app for replies, refunds and documents that still need follow-up. I independently designed and built the product, connected its paid features and shipped it on the App Store.
- Available now
- App Store
- Native app
- iOS
- Subscription & lifetime
- Pro