endorsedOngoing

Backend & NLP Platform for Hospitality POC to MVP

Services Covered:

Custom Software Development • AI and Machine Learning

Industry

Hospitality, Travel & Tourism

Duration

30 months

Budget

50K - 199K

Start Date

1 March 2018

Team Size

1-5

Engagement Model

Nearshore

Project Capability Score (PCS)

79/100

Good

Project Capability Score (PCS) estimates how capable Sciant 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

Hotel ResBot

Germany

Project Summary

Sciant assisted a hospitality company to move from a proof of concept to an MVP by delivering backend development and bespoke NLP processing, working in two-week sprints. Their work increased intent-recognition accuracy from under 5% to over 96% and demonstrated strong domain knowledge in NLP and hospitality integrations.

Key Challenges

Move from a proof of concept to an MVP and beyond
Build and train a bespoke NLP engine because commercially available NLP solutions were insufficient
Deliver backend development and hospitality integrations to meet the client's use case

Project Deliverables

Backend development to move from POC to MVP
Bespoke NLP engine development and training for intent recognition
AWS-based serverless infrastructure (Lambda and databases)
Multi-cloud integrations including Google Data Studio for analytics

Project Solution

Sciant provided expertise in NLP and hospitality integrations and worked in two-week sprints to move the client from POC to MVP and continue to build out the solution. They developed and trained a bespoke NLP engine because off-the-shelf solutions were insufficient, and implemented backend services primarily on AWS (Lambda and various databases) while leveraging a multi-cloud approach for analytics such as Google Data Studio. The Sciant team embedded with the client, supplying a project manager/scrum master, a senior full-stack developer, and an NLP expert, and brought in additional specialists as needed.

Project Outcome

Increased intent recognition accuracy from less than 5% to over 96%
Improved project management practices and strong team embedding with the client

Platforms

Cloud
AI/ML Platform

Tech Stack

Amazon Web Services (AWS)

Client Endorsement

Overall Review Rating

4.88

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“With Sciant's exceptional assistance, the company achieved 96% accuracy in their development process, a tremendous improvement from their previous less than five percent intent recognition accuracy. What really sets them apart is their knowledge about NLP capabilities and hospitality integrations.”
Brendan May
Managing Director
Date: 09/30/2020