NeuroSync

InspirationWorking as a paralegal at a personal injury law firm that specialized in traumatic brain injuries, I saw firsthand that healthcare scheduling is broken for TBI survivors. A significant portion of my daily job was simply trying to schedule medical appointments for our clients because the standard bureaucracy is nearly impossible for them to navigate alone. During recovery, executive dysfunction makes multi-step coordination exhausting. Patients face sensory overload that turns dense portals into a wall of noise , and standard medical jargon is unreadable mid-recovery. We realized that scheduling a doctor's appointment shouldn't feel harder than recovering from the injury itself. What it doesNeuroSync Health is a healthcare scheduling assistant designed specifically for traumatic brain injury recovery. The patient simply describes what they need in a chat interface that asks one question at a time, and NeuroSync handles everything between that sentence and the appointment. It features a low-stimulation dashboard to prevent cognitive overload. For modern clinics, it facilitates instant scheduling by sending a structured booking request to API-enabled providers. For legacy clinics without an API, an AI concierge steps in to draft emails, call the office, and keep the patient updated until the clinic replies. How we built itWe built the platform in a single day for the AWS Hacks 2026 event at Seattle University. The architecture begins with a stateless intake managed by AWS Lambda. We then utilize Amazon Bedrock to power our AI provider matching. The system routes requests through two paths: a live API for sub-second confirmation with modern clinics , and an email/voice human-in-the-loop system for traditional offices. We successfully built out the full UI and state end-to-end , featuring a fully working chat and dashboard flow. Challenges we ran intoA major challenge was balancing the strict technical requirements of medical coordination with the need for an extremely minimalist, low-stimulation user interface. Additionally, handling the dichotomy between modern API-enabled providers and legacy clinics required us to design a dual-path routing system. Due to the rapid one-day timeline of the hackathon , we had to mock several backend elements, which meant hardcoding 2 mock providers, simulating the confirmation emails, and using hardcoded scheduling responses for the demo. Accomplishments that we're proud ofWe are incredibly proud to have built a full end-to-end UI and working chat flow from scratch in just one day. Successfully integrating Amazon Bedrock to power our provider selection allowed us to prove our core thesis: that AI can genuinely remove the coordination burden from a TBI patient. Creating a system that seamlessly accommodates both modern API clinics and traditional legacy clinics via an AI concierge represents a massive step forward in patient accessibility. What we learnedWe learned that the true power of AI in the healthcare sector isn't just in diagnosis or data analysis, but in providing immediate administrative relief to vulnerable populations. We realized how critical it is to design for the specific cognitive loads of users, especially those dealing with sensory overload and executive dysfunction. On the technical side, building this application taught us how to rapidly and effectively integrate AWS Lambda and Amazon Bedrock into a cohesive, patient-first product architecture. What's next for NeuroSyncWe plan to build this out into a real product across two distinct upcoming phases. Phase II (Trust & Memory): We will implement a HIPAA-ready infrastructure featuring VPC isolation, least-privilege IAM access, and PHI encryption via AWS KMS. We will also develop a smarter data layer using DynamoDB patient records to track real-time scheduling state and remember patient preferences. Phase III (Reach & Care): We will focus on real healthcare integration with Epic, Cerner, and FHIR platforms to enable insurance-aware scheduling and automated clinic outreach. Finally, we will expand our AI capabilities to include plain-language medical message translation and appointment reminders.

Built With

• css
• javascript
• python