EndorsedendorsedCompleted

AI Engine Integration for Invoice Duplicate Detection

Services Covered:

Custom Software DevelopmentAI and Machine LearningAPI Development
Project hero — replace with case study imagery
Industry
Industry

Software Development

Duration
Duration

36 months

Budget
Budget

Confidential

Client Size
Start Date

1 January 2020

Team Size
Team Size

6-10 Employees

Engagement Model
Engagement Model

Nearshore

Project Capability Score (PCS)

gauge meter
77/100
Good

Project Capability Score (PCS) estimates how capable Qualysoft 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

Hyperfy

mapBucharest, Romania

Project Summary

Qualysoft expanded the client's internal development capacity and integrated an AI engine into a commercial product, using Python, TensorFlow, PyTorch, and C++. Their work streamlined the client's processes and reduced labor costs by about 60%, enabling broader deployment of the solution and receiving praise for professionalism and technical expertise.

Key Challenges

  • Expand the internal development team and add specialized AI expertise
  • Improve and develop an AI-based software product for invoice processing
  • Deliver a user-friendly product with a strong user experience

Project Deliverables

  • Integration of the artificial-intelligence engine into the final product
  • API integrations connecting different systems
  • Improved user interface and user experience
  • Deep-learning models and an ensemble of three neural networks for invoice duplicate detection

Project Solution

Qualysoft provided a dedicated team (6-10 employees) to integrate the AI engine into the client's product, connecting systems via APIs and improving the user interface. They implemented deep-learning solutions using Python, TensorFlow, PyTorch, and C++ and built a three-network ensemble to decide invoice duplicates. Their work enabled the product to be production-ready and supported deployment to enterprise customers.

Project Outcome

  • Reduced labor costs by 60% for the client's invoice payment process
  • Product deployed across a multinational banking and finance company's branches, improving efficiency and user satisfaction
  • Improved duplicate-invoice detection using an ensemble of three neural networks

Platforms

  • AI/ML PlatformAI/ML Platform

Tech Stack

  • PythonPython
  • TensorFlowTensorFlow
  • C++C++
  • RESTful APIRESTful API

Client Endorsement

Overall Review Rating

5star5 out of 5 stars

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

Qualysoft’s solution streamlined the client's processes and reduced labor costs by 60%, much to their delight. The team leveraged their extensive knowledge and technical expertise to deliver optimal solutions, and internal stakeholders praised the service provider's professionalism and integrity.

Cosmin Marinescu

Managing Partner & BDM

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

More Projects by Qualysoft

Completed
Software Development

EV Charger Software Development & Staff Augmentation

Preview of the EV Charger Software Development & Staff Augmentation
Budget:
Confidential
Duration:
24 months
Team Size:
Confidential

Services Covered:

+3 more