AI-Powered Message Automation Enhances Customer Engagement for Inunity
AI-Powered Message Automation Enhances Customer Engagement for Inunity
Client:
Region:
US
Industry:
Technology / Business
Inunity, a customer-first business texting platform, sought to enhance its SMS and web chat responsiveness by integrating generative AI capabilities. Partnering with SnapSoft, an AWS Advanced Tier Services Partner, Inunity launched a Minimal Viable Product (MVP) leveraging Amazon Bedrock and AWS-native tools to create personalized, automated message responses. The outcome was a scalable AI-driven solution validated in a secure AWS staging environment, enabling smarter, anonymized communication at scale.
Our partner said
About the Customer
Inunity is an acquisition-focused software company based in Nashville, Tennessee, with a portfolio of communication-centric products. Its flagship platform, TextLine, facilitates secure and scalable SMS-based communication between businesses and their customers.
Customer Challenge
Inunity needed to modernize its customer support workflows by introducing automation in its TextLine platform, which serves businesses managing thousands of customer interactions. The company faced several challenges, including the manual handling of large volumes of messages across SMS and web platforms, and the absence of automation tools capable of leveraging unique conversation histories and customer-specific knowledge bases. A critical requirement was implementing robust PII anonymization during data processing to ensure data integrity and meet security expectations.
Why AWS?
AWS was selected for its robust set of AI/ML services, scalability, and secure infrastructure. As an AWS Advanced Tier Services Partner, SnapSoft leveraged AWS-native tools and frameworks to ensure the MVP aligned with AWS Well-Architected Framework principles, delivering flexibility, performance, and cost-efficiency.
SnapSoft’s Contribution to the Solution
SnapSoft designed and deployed a minimum viable product (MVP) in a dedicated AWS account owned by Inunity to validate a Retrieval-Augmented Generation (RAG) approach tailored to the company’s historical SMS and web chat conversation data. The implementation unfolded in two phases. During the five-day Discovery phase, SnapSoft analyzed historical interactions, identified recurring communication patterns, and assessed the capabilities of Amazon Bedrock for text processing. In the subsequent 30-day Model Build phase, the team developed a data pipeline to process incoming messages using Bedrock-based text models. The MVP was validated using production-grade data in a secure staging environment. To ensure knowledge transfer and solution continuity, SnapSoft conducted a two-hour workshop demonstrating the deliverables and providing guidance to Inunity’s team.
AWS Services and Tools Used
- Amazon Bedrock
- AWS Lambda
- Amazon S3
- Amazon Aurora (Staging data source)
- AWS IAM (Access control)
- DynamoDB
- OpenSearch
- SQS
Results and Benefits
- Enhanced Automation: A validated MVP capable of automating SMS and web responses using AI, reducing manual workload.
- Scalability: Designed for future extension to speech-based interactions using AWS transcription and speech synthesis.
- Security & Compliance: Real-time PII anonymization pipeline reduced exposure risks and improved compliance readiness.
- Cost-Neutral Development: Entire engagement was AWS-funded, enabling innovation with no upfront investment from Inunity.
- Knowledge Transfer: Handover workshop ensured Inunity's teams were equipped to build upon the MVP.