endorsedCompleted

Android Virtual Assistant & ML Engine for Discussion Forums

Services Covered:

Mobile Application Development • AI and Machine Learning • Software QA and Testing

Industry

Artificial Intelligence and Machine Learning

3 months

Budget

Less than 10K

Start Date

1 November 2019

Team Size

1-5

Engagement Model

Onshore

Project Capability Score (PCS)

73/100

Good

Project Capability Score (PCS) estimates how capable Yugasa Software 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

Noida, India

Project Summary

Yugasa Software Labs developed an Android mobile app and underlying ML components (parser, data cleaner, ML engine) to organize and provide contextual answers from large discussion forums for a machine learning company. The vendor delivered a working proof of concept quickly and met the client's goals; the team integrated with the client during the engagement. The app was uploaded to the Play Store but later suspended due to data-scraping issues that the vendor had previously warned about.

Key Challenges

Make sense of very large amounts of unstructured forum data
Build a mobile virtual assistant that can answer contextual, layman-format questions
Navigate productization and compliance risks related to data scraping

Project Deliverables

Data parser
Small data cleaner engine
Machine learning engine to process forum data
Android frontend (mobile app / virtual assistant)
Testing and Play Store publishing

Project Solution

Yugasa produced a rapid proof of concept that scraped and clustered forum data, then built a small virtual assistant and an Android frontend. The vendor implemented a parser, data-cleaning pipeline, an ML engine for clustering/insights, delivered a limited-knowledge virtual assistant, performed testing and published the app to the Play Store. They also advised the client on practical productization risks.

Project Outcome

Demonstrated a working proof of concept within ~20 days
Delivered an app and ML components that met the client's goals
Uploaded the app to the Play Store (later suspended due to data-scraping concerns)

Platforms

Mobile

Tech Stack

Android

Client Endorsement

Overall Review Rating

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“The Yugasa Software Labs team developed a working proof of concept very quickly. The final delivered output achieved all of the goals that the company set out for it at the start of the partnership and performed up to their expectations. The team operated as part of the company for the project.”
Vivek Mittal
Staff Engineer
Date: 02/23/2021