What challenges did NewDay face before AI implementation?

Agents needed faster, more accurate retrieval of answers from knowledge articles during live calls. The team had to manage cost, latency, limited infrastructure, and low confidence in the initial proof of concept.

Source

ProductionEvidence: Medium50/100

NewDay builds a generative AI customer-service agent assist with over 90% accuracy

Use case typeCustomer service automationUpdated Jun 13, 2026

NewDay created a real-time generative AI assistant for contact center agents to answer customer questions from about 200 knowledge articles during live calls. The team used a serverless RAG design on AWS, iterated through eight experiment loops, improved accuracy from below 60% to over 90%, and cut answer retrieval time from 90 seconds to 4 seconds.

Organization: NewDay
Industry: Finance
Location: United Kingdom
Published: June 2025

Reported outcomes

Accuracy: More than 60% higher

Quality & accuracy

Accuracy: More than 90% higherTime: Approximately 90 secondsTime: Approximately 4 seconds

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Accuracy: 60% increase

AWS Machine Learning BlogJun 24, 2025Blog postInferred claimMedium evidence strength

Accuracy improved from below 60% to over 90%.

Normalized claim

Accuracy: 90% increase

AWS Machine Learning BlogJun 24, 2025Blog postInferred claimMedium evidence strength

Accuracy improved from below 60% to over 90%.

Normalized claim

Time: 90 seconds

AWS Machine Learning BlogJun 24, 2025Blog postInferred claimMedium evidence strength

Answer retrieval time fell from about 90 seconds to 4 seconds.

Normalized claim

Time: 4 seconds

AWS Machine Learning BlogJun 24, 2025Blog postInferred claimMedium evidence strength

Answer retrieval time fell from about 90 seconds to 4 seconds.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: NewDay
Provider: AWS
Maturity: Production
Linked source: AWS Machine Learning Blog

The solution rolled out to more than 150 agents and is planned for broader customer operations use

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1Agent Assist
2Customer Service Automation
3RAG

Built NewAssist as a real-time RAG chatbot/agent assist using Amazon Bedrock with Claude 3 Haiku.
Used Amazon API Gateway, Amazon OpenSearch Serverless, AWS Fargate, and AWS Lambda in a serverless architecture.
Created a golden dataset, custom data parsing for widget-based knowledge articles, and iterative evaluation gates before production.

Technologies

Amazon Bedrock Amazon API Gateway Amazon OpenSearch Serverless AWS Fargate AWS Lambda

Accuracy improved from below 60% to over 90%.
Answer retrieval time fell from about 90 seconds to 4 seconds.
Running cost was kept under $400 per month.

Architecture

NewDay built NewAssist as a real-time retrieval-augmented generation assistant for contact center agents. The system uses Amazon API Gateway, Amazon OpenSearch Serverless, AWS Fargate, AWS Lambda, and Amazon Bedrock with Claude 3 Haiku. The team created a golden dataset, customized parsing for widget-based knowledge articles, and used weekly human review plus pre-production evaluation gates to improve accuracy before rollout.

Sources & evidence1

Evidence: Medium50/100Evidence strength