SignalRX

Inspiration

Healthcare professionals are constantly making decisions based on clinical experience, but valuable peer perspectives are often scattered across conversations, communities, and disconnected resources.

I wanted to explore a simple question: What if a physician could ask a question and immediately understand how other healthcare professionals are thinking about it?

That idea became SignalRX an AI-powered workspace designed to turn physician perspectives into useful, understandable insights.

What it does

SignalRX lets users ask healthcare-related practice questions and explore perspectives from a network of healthcare professionals.

Instead of returning one generic AI answer, SignalRX:

Generates relevant peer perspectives Summarizes the collective response with AI Identifies common themes and signals Lets users inspect the underlying responses behind the summary Connects the insight back to the actual perspectives that produced it

The goal is to move from "What does AI think?" to "What are healthcare professionals saying, and what can we learn from it?"

How we built it

SignalRX was built as a modern web application using Next.js, React, TypeScript, and Tailwind CSS.

We created a synthetic network of healthcare professionals with different backgrounds and perspectives. When a user submits a question, SignalRX identifies relevant peer experiences and presents them through an interactive interface.

We also integrated AI to generate peer experiences and synthesize the responses into a concise network-level summary.

A major design goal was transparency: rather than hiding the AI's reasoning behind a single answer, users can inspect the source responses that contributed to the overall insight.

Challenges we ran into

One of our biggest challenges was making the AI output feel like a network of distinct healthcare professionals rather than several copies of the same chatbot answer.

We had to think carefully about:

Creating believable differences between professional perspectives Matching responses to the user's question Preventing the summary from overwhelming the underlying evidence Designing an interface that makes AI-generated insights easy to explore Keeping the prototype clearly separated from real clinical decision-making

Accomplishments that we're proud of

We're proud that we turned a relatively simple idea, asking AI a healthcare question into an interactive peer-intelligence experience.

Our biggest accomplishment is the connection between the three layers of the product:

Question → Peer Perspectives → Network Insight

Users can start with a question, see individual perspectives, understand the larger pattern, and then investigate the responses behind that pattern.

We're also proud that the prototype is fully functional as a web application.

Most importantly, we built SignalRX around the idea that AI should surface and organize perspectives rather than replace them.

What we learned

We learned that building a useful AI product is about much more than getting an LLM to generate a good response.

The hardest and most interesting part was designing the experience around the AI.

We learned the importance of:

Giving AI outputs meaningful context Showing the evidence behind generated insights Designing for exploration instead of one-shot answers Creating structured data that AI can reason over Building fallbacks so the core product remains usable Clearly communicating the limitations of synthetic and AI-generated information

We also learned how much more compelling an AI application becomes when the user can interact with the information instead of simply reading it.

What's next for SignalRX

Our next step is to evolve SignalRX from a peer-experience explorer into a physician intelligence and action platform.

Future versions could:

Identify recurring physician needs and concerns Detect emerging themes across conversations Connect insights to approved medical resources Generate follow-up workflows for medical-affairs teams Route unanswered scientific questions to the appropriate human expert Track whether a physician's question was ultimately resolved Integrate with existing healthcare and pharmaceutical workflows Expand beyond synthetic data to properly authorized, de-identified real-world feedback

Our long-term vision is simple:

Turn physician insights into meaningful action.

Built With

• next.js
• node.js
• react
• tailwind
• typescript
• vercel