From One-to-One Consultations to a 24/7 AI Styling Portal: How Loop8 Helped Styled by Angela Scale to Hundreds of Customers a Week

The Styled by Angela business model — personalized styling consultations rooted in body shape and color analysis — was constrained by the same ceiling that limits most one-to-one service businesses: every consultation required the founder's direct time, and the social-media content workflow that drives new customer acquisition was itself a manual job. Growing the customer base meant either spending more hours in consultations and content production, or finding a way to put the styling methodology into the hands of customers directly.

Loop8 LLC, an AWS Consulting Partner, designed and delivered an AI-powered virtual styling portal for Styled by Angela on Amazon Web Services. The engagement was structured as a four-week proof of concept that validated the core experience — photo upload, AI-powered body and skin-tone analysis, personalized recommendations, virtual try-on, and an admin caption generator — followed by a production deployment that scaled the same scope and integrated with Styled by Angela's existing booking flow (Calendly). The production architecture uses Amazon Bedrock for every AI capability, Amazon S3 for photo storage, Amazon Cognito for customer and admin authentication, AWS Lambda and Amazon API Gateway for serverless backend, and Amazon DynamoDB for customer profiles and saved looks.

Every Styled by Angela consultation required the founder's direct involvement: visually assessing body shape, evaluating skin tone and color season, building a recommendation list, and walking the customer through suggested looks. Customers booked through Calendly and were styled in a separate manual workflow. There was no through-line from the booking moment to a styling experience that could be delivered immediately, on the customer's schedule, without a second touch from the founder.

Customers received styling recommendations as written descriptions or as references to garments they had not yet seen on themselves. There was no way to render a customer in a recommended outfit before the customer purchased — leading to hesitation, returns, and customer questions that could only be answered by the founder.

Styled by Angela selected Loop8 LLC as their implementation partner for the combination of SMB focus, full-stack delivery, and proof-of-concept to production continuity. Loop8 delivered the complete solution — infrastructure-as-code, serverless backend, customer-facing portal, AI integration across four distinct capabilities, admin caption generator, and Calendly integration — as a single integrated engagement rather than requiring multiple vendors.

The same Loop8 team that delivered the four-week proof of concept delivered the production version. The architecture validated in the PoC moved into production with the same patterns scaled up — no rewrite, no relearning, no second-vendor handoff. All source code, infrastructure-as-code, configurations, and operational documentation were transferred to Styled by Angela at production cutover — the founder owns the platform, with Loop8 available as the ongoing partner of choice rather than a dependency.

Loop8 began with a four-week proof of concept that validated the core styling experience end-to-end. The PoC was structured to prove the AI capabilities worked at the depth Styled by Angela's customers needed — body-shape and color analysis, personalized recommendations, virtual try-on, and the admin caption generator — before any production commitment. The production deployment scaled the same scope and added the Calendly integration so prospective customers could move from booking to portal access through a single signed link.

Customer-Facing Portal

The portal is delivered through Amazon CloudFront for low-latency access across the United States. Customers sign in through Amazon Cognito, upload a full-body photo through a presigned Amazon S3 URL, and progress through analysis, recommendations, virtual try-on, and saved looks — all within a single browser-based session. The admin portal lives behind the same Cognito user pool with a separate permission boundary for the business owner.

AI Capabilities — All on Amazon Bedrock

Amazon Bedrock powers every AI capability: body-shape analysis (multimodal classification from an uploaded photo), skin-tone and color-palette analysis (multimodal vision identifies undertones and recommends a palette), personalized clothing recommendations (language reasoning translates the analysis into a specific recommendation list), virtual try-on (image generation renders the customer in selected garments so they can see the recommendation rather than imagine it), and social-media caption generation (language models produce platform-specific captions with hashtags and tone for the admin's content workflow). Each capability is invoked through a dedicated AWS Lambda function behind Amazon API Gateway, with the model tier chosen per use case to balance latency against output quality.

Data Tier and Customer Privacy

Amazon DynamoDB stores customer profiles, analysis results, and saved looks. Photo objects live in Amazon S3 with AWS KMS customer-managed encryption keys. The data tier is scoped per customer through Cognito user IDs flowing into IAM-scoped Lambda execution roles.

Booking Integration

The production deployment integrates with Calendly so prospective customers move from booking to portal access through a single signed link, eliminating the manual coordination step the founder previously absorbed. Third-party API tokens are stored in AWS Secrets Manager and injected into the Lambda execution environments at runtime — never committed to code or stored in environment variables in a UI.

Security and Compliance Baseline

AWS WAF sits in front of the API Gateway with a managed rule set covering common web exploits, rate limiting, and bot mitigation. AWS Certificate Manager terminates TLS at the edge. AWS Key Management Service provides customer-managed encryption keys for data at rest. AWS CloudTrail captures the API audit trail. Amazon CloudWatch ingests logs and metrics from every function and triggers alarms when latency, error rates, or throttling exceed thresholds.

Infrastructure as Code

The entire AWS environment is defined in Terraform with reusable modules covering the portal frontend, backend, AI integrations, data tier, security baseline, and integration plumbing. The infrastructure-as-code, source code, and runbooks were transferred to Styled by Angela at production cutover.