Automating the Conversation: The We Love Joe Success Story
The Client
We Love Joe is an all-in-one AI-powered customer communication platform designed to simplify and automate business interactions across multiple channels.
Unlike traditional virtual receptionists that only handle phone calls, this platform engages customers wherever they are – including SMS, web chat, and social media – ensuring a seamless and proactive experience 24/7.
By using advanced AI agents that learn from every interaction, businesses can automate appointment scheduling, answer complex inquiries, and even send payment links, all without adding extra staff. Ultimately, We Love Joe helps companies of all sizes reduce response times and increase sales by transforming their customer support from a manual task into a strategic, automated growth driver.
- Industry:Information Technology
- Company Size:2-10
- Country:France
It was honestly a great experience working with you and the whole team. Everything was super well organized, clear, and professional from start to finish. Everyone was really helpful throughout the project, which made the collaboration very smooth and enjoyable.
Challenges
- The Production Gap – The platform’s initial Next.js/ElysiaJS local environment lacked the scalability, high availability, and rock-solid reliability required for a commercial SaaS launch.
- Fragmented & Constrained Resources – The team needed to consolidate disjointed, off-cloud secondary solutions into a single, unified AWS environment while navigating early-stage budget limitations.
- The AI Architecture Dilemma – A critical decision point required choosing the most efficient path for their AI layer – specifically weighing the ease of managed models via Amazon Bedrock against migrating proprietary, in-house models onto Amazon SageMaker.
- Compliance & Market Barriers – To launch on the AWS Marketplace and sell to SMBs, the infrastructure required enterprise-grade security and strict GDPR compliance, including localized deployment within EU data residency boundaries.
To address these challenges, We Love Joe needed a functional migration VideoSignal from a local runtime to a unified, production-grade AWS environment. This architecture will optimize early-stage cloud spend, define a scalable path for the AI intelligence layer, and enforce localized EU compliance – establishing a secure, market-ready foundation for commercial SaaS expansion.
Solutions
Our solution established a high-performance, GDPR-compliant AWS foundation designed to host the new VideoSignal tool while preparing the infrastructure for the eventual migration of the entire “We Love Joe” platform.
At Cloudvisor we made use if a multi phase approach.
Milestone 1:
Production Infrastructure & AI Integration (Current Phase)
Serverless Containerization (ECS Fargate)
Migrating the backend to a fully automated, serverless environment to handle the Next.js/ElysiaJS stack without operational overhead.
Dual-AI Core (Bedrock & SageMaker)
Deploying a hybrid AI strategy - utilizing Amazon Bedrock for instant foundational model access (image analysis) alongside Amazon SageMaker for proprietary model training and inference.
Low-Latency Storage (DynamoDB & S3)
Implementing DynamoDB for lightning-fast metadata tracking and asynchronous state management, backed by encrypted Amazon S3 buckets to securely store and process incoming media assets.
CI/CD & Localized Security
Building a secure automated deployment pipeline with centralized credentials via AWS Secrets Manager, deployed entirely within EU boundaries for strict GDPR compliance.
Milestone 2:
Ecosystem Migration & Advanced Scaling (Future Phase)
Core Application Consolidation
Unifying the primary "We Love Joe" ecosystem by migrating all remaining off-cloud workloads into the new AWS environment.
Enterprise Expansion
Preparing the infrastructure for enterprise adoption by architecting multi-region availability and hybrid cloud capabilities.
Automated MLOps
Implementing continuous retraining loops for proprietary models on SageMaker to automate refinement based on live interactions.
AWS Services Used
- Amazon ECS (Fargate)
- Amazon ECR
- Amazon SQS
- Amazon CloudFront
- AWS GuardDuty & AWS CloudTrail
- Amazon S3
- Amazon Bedrock
- Amazon Route 53
- AWS Certificate Manager
Results
The project successfully transformed the platform’s local development environment into a high-performance, unified AWS architecture designed specifically for commercial SaaS scaling. This modern setup successfully positions the platform for enterprise readiness and immediate AWS Marketplace listing, ensuring the application remains highly available, cost-effective, and fully capable of handling complex, omnichannel customer communications at scale
Serverless ScaleEliminating manual server management
- Localized GDPR ComplianceSecured the environment within EU data residency boundaries and centralized credentials
-
Dual-Engine AI SovereigntyIntegrated Amazon Bedrock for instant, managed image analysis
-
Optimized Data & Asset LayerDeployed Amazon DynamoDB for lightning-fast metadata tracking
