MegaEdge Launch Recap | AI for Skilled Nursing is Live

"You don't actually care about AI."

That's how Megadata CEO and founder Shalom Reinman opened the MegaEdge launch webinar on October 8, 2026.

Operators aren't shopping for new technology because it's new. They care about rehospitalizations, census, labor costs, reimbursement accuracy, collections, and margin. So the question Megadata asked wasn't "what cool things can we build with AI?" It was "how can AI, sitting on top of all of your data, help you move those outcomes?"

MegaEdge is the answer to that question, and it's live now.

If you missed the session, here's the recap: what MegaEdge is, what the live demo showed, how security and accuracy work, and how to get started.

Watch the full replay: https://youtu.be/RUmOPrUzLvM

‍

Why AI for Skilled Nursing Has to Start With Outcomes

Shalom laid out the thesis behind MegaEdge in four parts. AI on top of your data should:

• Give you the answers you need immediately
• Let you use your data in the work you're already doing
• Tell you what you need to know, even when you didn't think to ask
• Keep important problems in focus until the data shows they're resolved

Today, getting value from data still takes real effort. You need to know which report holds the answer, what question to ask, and how to interpret what you see. Then you need to remember to come back and check whether things improved.

AI changes that math. Faster answers. Less time spent finding and analyzing information. More people across the organization able to use sophisticated data. Shalom summed up the long-term vision in three words: from visibility to execution. MegaEdge V1 is the first step, focused on getting you to answers faster.

What MegaEdge Is: An AI Layer on Nine Years of LTC Data

Here's the part that makes MegaEdge different from the AI tools flooding long-term care right now. For AI to give trustworthy answers, it needs a strong foundation underneath it:

• Clean, reliable data from across the organization
• Business logic that knows what the data means
• Security and permissions
• An understanding of how clinical, census, reimbursement, labor, expenses, billing, and financials relate to each other

That's what Megadata has spent nine years building with its customers. Today that foundation spans 2,600+ facilities in 42 states and 80+ live integrations, with LTC terms like census, HPPD, CMI, and EBITDAR already mapped to the right metrics.

"The AI is new," Shalom said. "But the foundation underneath is what we've built together with you."

So when this generation of AI became possible, Megadata didn't have to figure out how to get the data or what it means. The team could focus on putting intelligence on top of it.

Two ways in, one engine underneath

VP of Technology Itay Fried walked through how it works. At the center is the MegaEdge Engine, which understands not just your question but what you're actually concerned about. It identifies the data it needs (sometimes one source, sometimes several), pulls it from the Megadata foundation, and returns a clear, structured answer.

There are two ways to use it:

1. The MegaEdge platform. Designed for the everyday user. Ask a question about your facility or organization and get the answer with numbers, charts, context, and suggested follow-ups. You don't need to know where the data lives.
2. Connectors into the AI tools you already use. Bring your Megadata into Claude today, with ChatGPT coming soon. Your permissions come along automatically, the business logic stays intact, and setup takes minutes.

Why start with these two? Because customers asked for them. Dan Brody, Megadata's VP of Go-To-Market, said the team heard two requests again and again: "I need answers faster without being a report expert," and "My team already uses AI tools, so let Megadata work there."

Inside the MegaEdge Engine Demo: Tracing a Rehospitalization Spike

Instead of a scripted tour, the team ran questions attendees submitted when they registered. The demo used sample facility data, but the investigation path is one every clinical leader will recognize.

Step 1: Start with the big picture

The first question: "What is the rehospitalization rate across the company, and which facility has the highest?"

Before answering, the Engine asked a clarifying question about what exactly the user wanted to see. Then it returned a month-to-date answer. Yellow Terrace was running a 32% rehospitalization rate against a company average of 19%. The team pivoted to September with a simple follow-up, and the Engine went back to the same data and re-ran it for the full month.

Step 2: Rule out staffing

Next came a cross-model question: "Can you create a chart comparing rehospitalization rate and direct nursing HPPD to see if there's a correlation?"

This is where cross-model intelligence matters. The Engine pulled from clinical and labor data in the same answer. The chart showed the high-RTH buildings were adequately staffed. Staffing wasn't the culprit.

"Think about the time it would take to run a report like this," Dan said. A question that spans clinical and labor usually becomes a project involving several people across departments. Here, it took one question.

Step 3: Let the Engine point to the driver

So the team asked an open-ended question: "What other metrics could be driving the RTH percentage?"

This time they didn't tee up a specific answer. The Engine drew on its understanding of how LTC metrics relate and came back with clinical areas that stood out. Sorted by 30-day rehospitalization rate, Spruce River led at 21%, with 18.5 incidents and 9.7 falls per 1,000 resident days, both well above other facilities.

Step 4: Get granular

One more follow-up broke down the incidents at Spruce River over the last 30 days: 55 incidents, 30 of them falls and 17 of them skin tears.

In four questions, the team went from a company-wide number to a specific clinical pattern at a specific building. No report requests. No waiting on another department. For a deeper look at why this metric matters, see our guide to rehospitalization analysis in long-term care.

{{product-demo-call="/components"}}

Your Megadata, Now in Claude: A 10-KPI Leadership Report in Minutes

The connector demo showed what happens when your governed Megadata data lives inside Claude. You can ask the same kinds of questions as in the MegaEdge platform, but you also get everything else Claude can do with files, documents, and other connectors.

Shalom picked a task that eats hours at most LTC organizations: the leadership KPI report.

The old way

Someone at each building pulls numbers from several systems, fills out a spreadsheet, and emails it to corporate. Someone at corporate chases the missing emails, checks the numbers, consolidates everything, and builds the deck. Every week or every month, depending on the organization.

The MegaEdge way

Shalom gave Claude a blank KPI template covering 10 metrics across every facility and asked it to fill in September for a leadership meeting the next day.

Claude went through MegaEdge into each area of the business (census, payer mix, labor, clinical, reimbursement) and filled the template. Alongside it, Claude summarized the key issues for the meeting, including the labor problems and the buildings with the highest rehospitalization rates. Yellow Terrace and Spruce River showed up again.

Then Shalom asked Claude to color-code the sheet red, yellow, and green. Instantly the issues jumped out:

• Yellow Manor had a census problem, running at just 67%
• Spruce Lake and Maple Gardens were flagged for overtime and agency use
• PDPM rates varied noticeably from building to building

Making it weekly

Finally, Shalom used Claude Cowork's scheduling capabilities to set the report up as a recurring task: a fresh Excel file each week, covering the last seven days. Every week, Claude pulls the numbers through MegaEdge and delivers the report.

That's one example. With the connector, you can also combine Megadata information with email, calendar, and other connected tools, analyze it alongside your own documents, and build it into broader AI workflows your team designs.

AI Security, Permissions, and Accuracy: Questions Answered Live

Itay addressed the biggest concern about AI in long-term care head on: security. Many off-the-shelf AI tools leave security for the customer to figure out. MegaEdge carries it across the full stack, from the data foundation through the Engine to the AI tools. The Megadata foundation is SOC 2 compliant, and Megadata remains committed to HIPAA compliance, especially as it relates to AI and PHI. Existing Megadata permissions and security carry over to MegaEdge with no extra setup.

The live Q&A went deeper on three questions.

What if my Claude isn't HIPAA compliant?

You can get a HIPAA-compliant version of Claude through Claude Enterprise or by connecting Claude through Amazon Bedrock. If your Claude setup isn't HIPAA compliant yet, MegaEdge withholds resident-level detail and still provides clinical and other data in aggregate form. Once your Claude is compliant, resident-level information can flow through.

How do I know the numbers are right? Doesn't AI make things up?

No AI model is perfectly deterministic, Itay explained. That's exactly why the MegaEdge Engine sits between the AI tool and your data. Connect Claude straight to raw data and you're leaving it to Claude to pick the right sources and calculations. With MegaEdge in the middle, the Engine selects the right data sources, handles the calculations, and runs structured queries against Megadata's governed data. The team tested answers extensively with industry experts before launch.

Can someone see data for a building they don't have access to?

No. Some AI tools handle security with guardrails, essentially telling the AI what not to do. MegaEdge doesn't rely on that. Access is based on the user, exactly as it is in Megadata. If an administrator only has access to their own facility, that's the only data the AI can reach. Ask the question any way you like; there's no other data available to return. The AI interprets the question and explains the answer, but a traditional permission architecture sits underneath, and the AI can't get around it.

Where MegaEdge Is Headed Next

"What we're releasing today is only the beginning," Shalom said.

Our vision for where we're taking MegaEdge includes AI-powered dashboards and reports built from the charts the Engine already generates, more intelligence tailored to each person's role, and a mobile experience. We're not slowing down, and we'll keep adding more tools as the platform grows.

For now, the focus is simple: get skilled nursing leaders to trustworthy answers faster, wherever they work. If you're curious how Megadata thinks about practical AI in LTC, our realistic look at AI in long-term care is a good companion read.

How to Get Started With MegaEdge

The launch webinar made three things clear about AI for skilled nursing done right:

1. The foundation matters more than the model. MegaEdge works because it sits on nine years of governed, LTC-specific data.
2. Cross-model questions are where the time savings live. Clinical tied to labor, census tied to finance, answered in one conversation.
3. Your data should work where you work. The Claude connector brings Megadata into the reports, files, and workflows your team already runs.

Already a Megadata customer? 30-day free trials roll out in batches over the next few weeks. Since your data is already connected, there's nothing to set up. Your CSM will reach out about a week before your activation.

Not using Megadata yet? Book a call with our team to see how MegaEdge and the Megadata foundation can work for your organization.

Megadata brought your data together. MegaEdge helps you ask it anything.

Ask. Analyze. Act. Welcome to MegaEdge.

‍