endorsedCompleted

AI Platform Development for Development Bank

Services Covered:

AI and Machine Learning • Website Development • Solution Architecture

Industry

Financial Services

Duration

3 months

Budget

10K - 49K

Start Date

1 October 2015

Team Size

1-5 Employees

Engagement Model

Onshore

Project Capability Score (PCS)

72/100

Good

Project Capability Score (PCS) estimates how capable ThirdEye Data 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

Confidential

Washington, United States of America

Project Summary

ThirdEye Data developed a proof-of-concept AI platform for a multilateral development bank to search and identify experts across large volumes of institutional documentation. The solution used IBM technologies (Watson, Web Explorer), NLP techniques, and Python/Java development to crawl, analyze, and surface ranked results in a natural-language frontend. The POC succeeded and the bank approved additional funding to scale the project.

Key Challenges

Need to analyze institutional documentation to identify experts by curriculum, past work, and publications.
Difficulty locating people with prior project experience across regions from disparate reports and web content.

Project Deliverables

Data crawlers and ingestion pipelines to index project completion reports
NLP-based processing using ontologies, metadata, and machine learning
Ranking/rating system to score team leads and expertise
Natural-language search frontend application to find experts by country, sector, and experience
Technical architecture documentation showing component interactions and integrations

Project Solution

ThirdEye Data crawled and indexed project reports and institutional documents using IBM technologies (Watson and Web Explorer), then applied ontologies, metadata extraction, NLP, and machine-learning methods to rate and identify experts. They built a frontend natural-language search application enabling queries by country, sector, and years of experience, and delivered a full technical architecture detailing component integrations. The team implemented Python and Java code where needed and held weekly progress reviews.

Project Outcome

Successful proof of concept that demonstrated expected results and led to approval and plans to scale up the project.
Incorporated Python and IBM technologies to address gaps in existing tools.
Delivered full technical architecture and maintained weekly check-ins to manage progress and adjustments.

Platforms

Tech Stack

Python
Java
Amazon Web Services (AWS)
Azure

Client Endorsement

Overall Review Rating

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“The application was created as a proof of concept in order to test the technology, producing solid results and leading to plans to scale it up. The team responded well to changing technology needs, incorporating Python as required, while always keeping on top of the project's weekly progress.”
Anonymous
IT Lead Specialist
Date: 06/01/2018