Meet Your Next Doctor, AI

The health industry is going to flip on its head in the next decade

I’ll share which AI tools I’m exploring, some experiments I’m conducting, and insightful information about what I’m observing in the world every week.

🔧 Three Tools I’m Testing

🦖 Datasaur - A deep tool for building RAGs and fine-tuned models for your business use cases. I’ve only gotten through building and testing RAG (Knowledge bases) in their system. I’ll be working to fine-tune a model at some point.

👨‍🔬 Perplexity Deep Research - One of the latest AI tools to add a deep research capability. I found its research to be impressive, but it was a bit brief in its report. Google’s and Grok 3 (brand new) responses are more complete.

🎥 Klap - A service that will help make short hook videos from longer videos. You point it to a YouTube video, and it analyzes it and then creates several shorts. The results were not great when I tried it on one of our webinar videos. I bet it’d work well for a video podcast or interview format.

🧪 AI Experiment of The Week

This week, I’ve been working with Cursor as I continue to explore the AI coding tools and the new AI development stack. Cursor is an AI coding development IDE with a chatbot and agent (Composer) to help create software. There’s great content for teaching you about using Cursor to build applications.

However, I wanted to use Cursor to expand on an application I’ve been working on in Replit. While Replit has been fantastic for adding features and building the structure of my application, it can be stubborn when debugging and fixing specific issues. Cursor is also great for making styling changes quickly and adjusting the layout simply. I followed this video to connect the two.

Now, you can use Cursor on it’s own to create applications from stratch, as well. But, with Replits built in server, database, and deployment options it makes the combined stack very formidable.

I’m going to continue to work with Cursor in three ways (solo, with Replit, and with Vercel v0). My long term goal is build a robust AI development stack that I could recommend to begineers and experts. At the moment, Cursor will definitely have a spot in the stack.

📰 Article of The Week

Google Introduces Co-Scientist Agent Systems - Science and medicine are rapidly changing and AI is accelerating the pace of learning and experiementation. As Google put it in the annoucement:

We introduce AI co-scientist, a multi-agent AI system built with Gemini 2.0 as a virtual scientific collaborator to help scientists generate novel hypotheses and research proposals, and to accelerate the clock speed of scientific and biomedical discoveries.

The co-scientist was tested against several highly technical and difficult areas like drug repurposing, identifying novel treatement targets, and how antimicrobial resistance evolves. You can dive into the research paper to really get deep on these use cases and how they’ve developed the new agent systems for co-scientist.

The AI co-scientist continuously generates, reviews, debates, and improves research hypotheses and proposals toward the research goal provided by the scientist.

Towards an AI Co-Scientist

This is gets out of my wheelhouse pretty quick but shows the profound changes happening in the scientific research and medical spaces.

🌎 Where the World is Going

Confined to bed with the flu this week, I found myself contemplating how AI is reshaping healthcare – not in some distant future, but right now. While traditional medical diagnosis relies on a doctor's experience with thousands of cases, we're rapidly approaching a world where AI systems can analyze millions of cases, symptoms, and outcomes simultaneously. This isn't about replacing doctors; it's about augmenting their capabilities with unprecedented precision.

What fascinates me most is how AI could transform the patient experience through continuous, contextual health monitoring. Imagine your health data – from breathing patterns and heart rate to nutrition and sleep cycles – being analyzed alongside environmental factors, local health trends, and your personal medical history. Rather than waiting until you're sick enough to visit a doctor, AI could spot potential health issues before they become serious, suggesting preventive measures or early interventions. My recent bout with the flu made me realize how valuable early detection could be, potentially shortening recovery times and reducing the spread of infectious diseases.

But the real breakthrough isn't just in diagnosis – it's in democratizing medical expertise. While leading specialists are concentrated in major medical centers, AI has the potential to bring that level of diagnostic capability to any clinic, anywhere in the world. As someone who's seen both the strengths and limitations of our current healthcare system this week, I'm convinced that AI's role in healthcare isn't just about efficiency – it's about fundamentally improving access to quality medical care for everyone. That's a future worth getting excited about, even when you're stuck in bed with the flu.

👨‍💻 About Me

Just a Guy with An Ostrich

My name is Charlie Key. I love technology, building awesome stuff, and learning. I’ve built several software companies over the last twenty-plus years.

I’ve written this newsletter to help inspire and teach folks about AI. I hope you enjoy it.

➡️ Learn More About The Guy ⬅️