EndorsedendorsedCompleted

React Front-End for OCR Labeling Tool on Azure

Services Covered:

Custom Software DevelopmentWebsite Development
Project hero — replace with case study imagery
Industry
Industry

Information Technology

Duration
Duration

4 months

Budget
Budget

Confidential

Client Size
Start Date

1 April 2023

Team Size
Team Size

1-5 Employees

Engagement Model
Engagement Model

Nearshore

Project Capability Score (PCS)

gauge meter
79/100
Good

Project Capability Score (PCS) estimates how capable Limestone Digital is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS


Client

Synatic

mapJohannesburg, South Africa

Project Summary

An IT services company (Synatic) hired Limestone Digital to build a React-based website to assist with labeling PDF pages and images for an OCR system, extending an open-source Microsoft labeling tool and integrating it with the client's API. The vendor delivered a front-end that maintained document page hierarchies, supported file upload/delete, managed fields per document/file, and provided a labeling interface. The client reported strong project management, frequent demos and standups, and high responsiveness from the team.

Key Challenges

  • Develop a website to assist with labelling PDF pages and images for an OCR system.
  • Extend Microsoft's open-source labeling tool to call the client's API instead of the limited prebuilt API and add additional functionality.

Project Deliverables

  • New React front-end for labeling custom AI models on Azure Cognitive Services.
  • Updated open-source labeling tool to call the client's API instead of the limited prebuilt API.
  • Document page hierarchy management.
  • File upload and deletion functionality for example files.
  • Field management per document and per uploaded file.
  • Labeling interface to tag documents based on defined fields.

Project Solution

Built a new React front-end to handle labeling for the client's custom AI models using Azure Cognitive Services. The team modified and extended an open-source Microsoft labeling tool to call the client's API (replacing the prebuilt API) and implemented features to maintain a document page hierarchy, upload/delete example files, manage fields per document and file, and label documents based on those fields.

Project Outcome

  • Project delivered on time with superb project management and high responsiveness.
  • Positive client feedback: the team was friendly, highly knowledgeable, provided prompt feedback, and worked overtime to complete the project.

Platforms

  • WebWeb

Tech Stack

  • React.jsReact.js
  • AzureAzure

Client Endorsement

Overall Review Rating

4.63star5 out of 5 stars

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

Limestone Digital’s work received positive feedback from the client. Daily stand-ups, weekly demo sessions, and ad-hoc sessions on Slack ensured smooth project implementation. The team was friendly and highly knowledgeable. Their responsiveness and prompt feedback were noteworthy.

Ryan Kotzen

Technical Lead

Note: This endorsement is based on publicly available client feedback from external review sources.

More Projects by Limestone Digital

Ongoing
Information Technology

Frontend Staff Augmentation for Advertising Management Software

Preview of the Frontend Staff Augmentation for Advertising Management Software
Budget:
Confidential
Duration:
84 months
Team Size:
1-5

Services Covered:

+1 more
Ongoing
Software Development

Event Management Platform & SendGrid Integration

Preview of the Event Management Platform & SendGrid Integration
Budget:
50K - 199K
Duration:
105 months
Team Size:
1-5

Services Covered:

+1 more