Caylent Catalysts™
Generative AI Strategy
Accelerate your generative AI initiatives with ideation sessions for use case prioritization, foundation model selection, and an assessment of your data landscape and organizational readiness.
Mediaura Fast-Tracks GenAI Healthcare Product to Commercial Readiness in 45 Days
Generative AI
At a glance
Mediaura, a full service digital marketing agency, partnered with Caylent and AWS to build Patient Health Lens – an AI-native application that helps patients better understand their health data and prepare for more effective conversations with their providers.
Solution Implemented
→ Designed and delivered the Patient Health Lens application in just 6 weeks, moving from ideation to a demo-ready product.
→ Implemented secure document upload and depersonalization workflow for lab results, diagnoses, and medication lists—avoiding protected health information (PHI) exposure to the model or underlying infrastructure.
→ Built a GenAI chat experience powered by Amazon Bedrock and Claude 3.7 Sonnet, with a proprietary medical knowledge base and cross-referencing capabilities.
→ Delivered a production-grade architecture using Amazon Cognito, Amazon OpenSearch Serverless, Amazon S3, Amazon DynamoDB, AWS CloudFront, Python FastAPI, React, Terraform, and WebSocket.
Outcomes Expected
→ Delivered a production-ready GenAI application in less than 45 days, enabling real-world validation of the product concept.
→ Shifted stakeholder conversations from speculative concept to demonstrated capability, generating early engagement from large healthcare partners.
→ Positioned Mediaura for commercialization with an estimated reach of 3M+ B2B users and 5M direct-to-consumer users, establishing a scalable foundation for future growth.
Empowering Healthcare Patients Through Generative AI
Mediaura is a full-service digital marketing agency with deep experience in healthcare. After more than two decades of custom software and marketing work, the company now emphasizes digital marketing services while selectively investing in AI-driven products that can deepen relationships with its healthcare clients.
Founder and CEO Andrew Aebersold is a seasoned technologist and inventor with previous software patents. Patient Health Lens is a GenAI product concept he incubated inside Mediaura, with the intention of spinning it into its own entity once the business case and commercialization path are clear.
Business Challenges
Aebersold had seen the same problem repeatedly in his everyday life. “I kept having these personal experiences of not having certain information readily available when dealing with healthcare matters for myself and loved ones,” said Aebersold. “We wanted to develop a way to empower end users and provide the data needed for greater honesty and transparency going into meetings with healthcare providers.”
Acknowledging that people are overwhelmed by lab results, diagnoses, medications, and underprepared for short, rushed provider visits, Aebersold noted they often struggle with medical terminology, forget key details once they’re in the exam room, and leave unsure of what to ask next.
He believed GenAI could help patients pull all information into a clear, human story to share with their providers. But there were two big obstacles. First, time. The GenAI space moves at a blistering pace. A traditional six-month build could mean that by the time Mediaura had something to demo, competitors might already be in the market, models would have changed, and budget spent could go to waste. “I needed to know quickly whether Patient Health Lens had traction with health systems, investors, and partners,” said Aebersold.
Second, risk and complexity. Healthcare data brings PHI and HIPAA considerations, and generic consumer AI tools are not designed with those requirements in mind. Standing up the right architecture, choosing the right models, and enforcing safety and tone controls would demand a level of expertise that Mediaura didn’t want to slowly assemble in-house. Doing it themselves or working with a firm that needed months to ramp up would erase the speed advantage they were looking for.
“Throughout the project timeline, Caylent remained consistent and professional. I could focus on my day-to-day tasks and didn’t have to worry about project oversight or monitoring timelines. This engagement was the opposite of previous experiences I've had with consultants.”
Andrew Aebersold
Founder and CEO
Solution
Mediaura engaged Caylent through the AWS Generative AI Innovation Center to design and deliver the Patient Health Lens chatbot in six-weeks. Caylent’s combination of AWS expertise, proven GenAI experience, and collaborative development fit the fast-paced startup needs.
The goal: to get a real, working product developed as quickly as possible so Aebersold could show it to health systems, advisors, and investors, gather feedback, and make informed decisions about commercialization and next-stage investment.
The resulting solution lets patients sign in securely, answer guided questions about their health, upload and depersonalize medical documents, and then chat with an AI assistant that summarizes their information in concise, empathetic language. The assistant is grounded in vetted medical knowledge based on a proprietary blend of data rather than the open web and is explicitly designed to support better conversations—not to diagnose or prescribe. The architecture was intentionally designed to support production use from day one, minimizing regulatory risk while enabling real-world evaluation.
From Idea to Proof-of-Value in 45 Days
Caylent delivered a production-ready GenAI application in just 45 days, providing Mediaura with tangible proof of product viability. The team kept scope tight, focused on high-value features, and consistently drove toward the goal of having something real that could be demoed quickly.
“Throughout the project timeline, Caylent remained consistent and professional. I could focus on my day-to-day tasks and didn’t have to worry about project oversight or monitoring timelines, said Aebersold. “This engagement was the opposite of previous experiences I've had with consultants.”
For Aebersold, accuracy and speed were critical. A six-month timeline would likely have stalled momentum or allowed competitors to move first. Instead, he now has a functional product ready to present to customers, advisors, and investors. That changes the nature of the conversation: from “What if we could build this?” to “Here’s what we’ve already built and how should we tailor it for your organization?”
Early discussions with healthcare partners and potential investors are underway, and Mediaura is actively evaluating commercialization paths based on real usage feedback rather than assumptions, reducing uncertainty and strengthening investor confidence.
Technologies Used
Caylent partnered with Mediaura’s team to develop Patient Health Lens using Amazon Bedrock with the Claude 3.7 Sonnet model for conversational understanding, summarization, and real-time patient support. The solution implemented secure access through Amazon Cognito and enabled medical document ingestion and analysis with Amazon S3, Amazon DynamoDB, and Amazon OpenSearch Serverless to store, retrieve, and surface relevant patient information during chat. The application experience was delivered through a React front end and a Python FastAPI backend, with real-time interaction powered by WebSocket connectivity. To support scalable, production-ready delivery, the platform leveraged AWS CloudFront for performant content distribution and Terraform for consistent, automated infrastructure provisioning across environments.
