Activity
880 followers
-
Frank Rydzewski shared thisIt's been a little while since my last post about using Claude Code for work and other fun projects. I thought I'd blog about my latest, most personal use of these advanced technologies. It's still somewhat of a work in progress, and there is another trove of letters we believe are at a distant cousin's home, but I figured people would enjoy seeing applications of agentic AI being used outside of just "writing software." https://lnkd.in/gRKwuxCr
-
Frank Rydzewski posted thisWe're now part of ChartSpan Medical Technologies, the largest Chronic Care Management (CCM) and Annual Wellness Visit solution provider in the US. It's exciting to be bringing the products we built at Validic to our own clinical staff, to augment their CCM programs with data from our patients' homes. I can't wait to start seeing the positive outcome we have on the lives of our patients and their families. Stay tuned to watch us grow together!
-
Frank Rydzewski shared thisIn late January, I posted a company-wide memo about my experience and thoughts around agentic AI and its impact on software engineering. Given that this was ended up being a direct influence on our decision to pause work for two weeks and empower our workforce, it seemed relevant to share alongside my recent posts about how these new technologies are influencing and transforming our thinking. https://lnkd.in/ghtvYtRQ
-
Frank Rydzewski reposted thisFrank Rydzewski reposted thisFor the 7th consecutive year, Validic has been recognized by the MedTech Breakthrough Awards, this time winning "Best IoT Healthcare Platform" for Validic Inform. 🏆 This milestone means a lot to us. It reflects the work our team puts in every day to make personal health data more accessible, actionable, and interoperable for health systems, health plans, life sciences organizations, and developers building the future of care. Validic Inform connects data from 700+ devices and health apps through a single API. No per-device builds, no redundant integrations. Just clean, structured health data that powers earlier interventions, more personalized care plans, and stronger patient engagement. And we're just getting started. We've opened the platform to a new generation of builders. Developers and healthcare innovators can now self-sign up through our developer dashboard to immediately access the API and start building. As healthcare enters the agentic era, AI systems need a trusted, real-time health data foundation, and that's exactly what we're built to be. To our customers, partners, and the patients whose health outcomes drive everything we do: thank you. Read more in our press release: https://lnkd.in/drvwsFbf #DigitalHealth #HealthTech #MedTech #HealthcareInnovation #IoT #RemotePatientMonitoring #MedTechBreakthrough
-
Frank Rydzewski reposted thisFrank Rydzewski reposted thisA completely normal developer experience when building a health app: Week 1: "I'll add Fitbit integration, can't be that hard." Week 3: Still reading Fitbit's OAuth docs Week 5: Fitbit works! Adding Apple Health next Week 7: Apple Health works differently on iOS 16 vs 17. Week 9: Adding Garmin Week 11: Garmin deprecated that endpoint Week 13: Someone asks, "Can we add Withings?" We've been there. Actually, we've been there for 13 years, which is why we built Validic Inform. 700+ integrations. One API. We handle the endpoint deprecations, the SDK updates, the manufacturer quirks. You get clean, normalized health data and your sanity back. And now you can sign up and start building in minutes, not months: https://lnkd.in/gv8pEnDf #Developers #HealthTech #BuildingInPublic #DigitalHealth #DevHumor
-
Frank Rydzewski shared thisWe've been doing a lot of work to make it easy to just "get building" and see what you can do with personal health data. Today marks the beginning of our mission to release features targeted at making it easier for you to make your ideas real. We can't wait to see what you build.Frank Rydzewski shared thisToday, we're releasing two new capabilities that change how developers and builders access Validic Inform: a developer dashboard and self-sign up. No contract. No credit card. No sales call. Just instant access to the same personal health data infrastructure that the top U.S. health organizations have depended on for over a decade. 700+ device integrations, one normalized API, and HIPAA-compliant infrastructure from day one. Whether you're a solo developer with an idea or a team ready to ship, you can now sign up and start building with real health data infrastructure today. Learn more: https://lnkd.in/gWZRTkga #DigitalHealth #HealthTech #Developers #HealthcareInnovation #Validic
-
Frank Rydzewski shared thisAt Validic, we acknowledge that there are a lot more builders now than there were a year ago. Agentic AI did that, and the people building aren't all developers in the traditional sense. We paused our pitch cycle for two weeks so our teams could go deep on these tools and internalize that shift. What came out of it wasn't just better AI literacy. It was a clearer view of where our products need to go. We're evolving Validic to serve traditional developers using agentic AI and the new generation of builders behind them. Both are real. Both are here now. Why we think it matters: https://lnkd.in/gMswVb2TEveryone's a Builder Now: How the Agentic Age Is Reshaping Health Data DevelopmentEveryone's a Builder Now: How the Agentic Age Is Reshaping Health Data Development
-
Frank Rydzewski shared thisI've talked a lot about agentic AI and how I've been adopting it into my personal and professional life. At Validic, we've continued our journey in adopting these tools and making them available to all roles. I'm observing some interesting organizational side-effects in our journey and blogged about it. I think it's important to understand the impact of having a fragmented or unaligned understanding of these tools and what they can do. Regardless if you decide to use them or not use them, it's important to have that unified view across all roles (minimally leadership), and not just confined to product and engineering. https://lnkd.in/gQDBxFpmConway's Law in Reverse: How a Two-Week Pitch Pause Rewired Our Org's ThinkingConway's Law in Reverse: How a Two-Week Pitch Pause Rewired Our Org's Thinking
-
Frank Rydzewski shared thisI've been meaning to get around to posting about something we decided to do at Validic in February. We decided to ask our product and engineering teams to stop working for two weeks and spend that time learning and exploring Agentic AI code development tools and techniques. It turned out to be one of the most fun two weeks I've had in my entire career. Read more here: https://lnkd.in/gR-jwKc2 If anyone is early in their agentic journey and wants to talk, hit me up. I've "claude whispered" quite a few colleagues and former colleagues over the past couple months, and got to witness their "gateway" moment.
-
Frank Rydzewski reacted on thisFrank Rydzewski reacted on this86% of physicians review patient wearable data, yet only 6% have it in their workflow. I've spent 13 years on this. Here's what's really going on. The data comes from a new AMA survey (link in comments). "Only 6% in the workflow!" we all cry. It must be interoperability ... blame the EHRs! Wrong. It must be reimbursement ... blame the payers! Closer, but wrong. I've sat in the rooms where these decisions are made. What really kills these initiatives? The calendar. Health plans are re-bid every year. Hospitals fight to preserve their shrinking operating budgets annually. So every investment cycle is viewed through the lens of in-year ROI. It goes like this: Health system: "Wearable data will deliver better outcomes and lower costs over time. The payer should cover it." Payer: "We can't pay for wearables, we won't see the ROI this year. The health system should cover it as part of its mission." Both: "We know how to diagnose/pay for diseases and perform surgeries. Let's stay in our comfort zone." Healthy behavior change (actual "health care") doesn't solve this equation. A patient slowly walking themselves back from prediabetes over three years is invisible to a model that closes the books in December. So the programs don't pencil out. They don't get covered, the data doesn't get built into the workflow, and next year another survey rediscovers the same 6%. Meanwhile, as patients, we keep scratching our heads, saying, "This is so obvious, why can't my physician see my health data?" We keep treating this as a data problem or an integration problem because those have vendors and roadmaps attached. This is a time-horizon problem. The calendar just sits there, structural, deciding everything downstream while we argue at the edges. The data was never the hard part. The calendar is.
-
Frank Rydzewski reacted on thisThis chapter was a huge labor of love, and I'm very proud of what we pulled together.Frank Rydzewski reacted on thisChapter 9 of the upcoming SRE Book (2nd Edition) is now available for Early Release readers on O'Reilly! Complex systems fail—it’s an unavoidable reality of software engineering at scale. But how we design our on-call rotations and incident management frameworks makes all the difference between a minor blip and a prolonged outage. I am incredibly proud to be a co-author of Chapter 9: Incident Management and On-call, alongside David Huska, Garrett Plasky, Jennifer Mace, and Michelle Brush. As part of this chapter, I wrote the section on Potential Ways of Structuring On-call. Designing a sustainable on-call rotation is a delicate balancing act. In this section, I dive deep into the core constraints and inputs that engineering leaders must navigate to protect team health while maintaining system reliability: Optimal Paging Load: Applying Google's "Treynor Limit" to prevent alert fatigue and ensure responders actually have the cognitive capacity to learn from incidents. Hours of Availability: Weighing the real business risks against the high organizational and human cost of 24/7 schedules. Response Time Dynamics: How strict response requirements (like 3-to-5-minute SLA windows) shift engineering lifestyle, commutes, and require explicit handoff planning. System & Customer Constraints: Navigating everything from localized labor regulations to highly specific regional, citizenship, or compliance requirements from enterprise customers. The chapter also explores how we approach incident management at Google (including our IMAG framework), managing the human element under high pressure, a brilliant case study from Etsy, and the evolving role of AI and autonomous agents in production safety. Check it out here: https://lnkd.in/eXjdTpX7
-
Frank Rydzewski reacted on thisFrank Rydzewski reacted on thisWith our acquisition of Validic, ChartSpan can now offer integrated Chronic Care Management with Remote Patient Monitoring, ensuring your patient’s remote health data can be transformed into clinical action. Join us on July 16 at 1 pm ET for a practical conversation about what the combined program looks like, including: ✅ How real-time device data helps the ChartSpan care team catch issues earlier and take clinical action between visits ✅ What a full-service partner handles on your behalf, including device logistics, enrollment, services, and billing ✅ How one trusted clinical team can support two integrated programs Sign up today to learn how we employ real-time health data to offer more personalized patient support.
-
Frank Rydzewski reacted on thisFrank Rydzewski reacted on thisIt's been a rough week. I received the call last Tuesday morning, shortly after dropping my youngest off at school. You had been in a motorcycle accident on the way to work, and you didn’t survive. I didn’t yet know exactly where it had happened. As I drove, I passed the intersection and saw your helmet on the ground. It was the same helmet I had seen so many times over the past five months, sitting on your desk while you worked alongside our team. You joined us as an intern in January, though the team could see your potential from the moment they first interviewed you. You were intelligent, hardworking, kind, and a quick learner. In just a few months, you connected with so many members of our community and supported them in their use of technology. We had just created this intern position, and you were among the first to step into it. We were excited to invest in someone early in their career, to give you real experience, and to create a place where you could learn, contribute, and start figuring out what might come next. Jordan, you made the most of that opportunity and, even in just a few months, you had become a highly valued member of our department. Knowing you were still at the beginning of that journey is part of what makes this so hard. We were lucky to know you, work with you, watch you grow, and have you as part of our team, even for far too short a time. You set the bar high for those who will follow, and every new intern we welcome will remind us of the promise we saw in you. https://lnkd.in/g7X2PsUR
Experience
Education
-
-
-
-
-
-
-
Licenses & Certifications
Organizations
-
KC Nanobrews
Chief Web Technologist, Event Organizer, Home-brew Evangelist
Other similar profiles
Explore more posts
-
Deepthi Bathina
EHRs have ALWAYS been software companies focused on electronic health records – and they will continue to be. That’s their DNA. Pause and think - If the goal is for clinicians to deliver hyper-personalized care – care that is both clinically and financially significant – shouldn’t it be pre-validated by real physicians and certified coders, built on intelligence that goes way beyond just one EHR’s data?? Not handed off to already strained health systems to stitch together… or to already burnt-out clinicians to validate over years while costing the system a lot. Meanwhile, clinicians are drowning, health systems are financially struggling – and yet you’re told to keep waiting on the next big roadmap promise from EHRs. That’s why RhythmX AI fuses all relevant data – clinically + financially validated at every layer – deeply integrated into your EHRs, instead of giving you a costly toolkit to figure out on your own. ⸻ Here’s how RhythmX AI stands apart: 🔹 8+ diverse data sources integrated today (EHR + non-EHR, health plans, formularies, social data) – interoperable to factor in hundreds of your own pathways into whole-patient intelligence. 🔹 296,000+ hyper-personalized treatment recommendations already delivered into EHR workflows – with every dismissed one immediately reviewed + iterated by RhythmX AI physicians. 🔹 25,000+ clinical validations by 30+ practicing physicians already completed – with certified coders ensuring medical economics rigor. 🔹 300M patients, 4.4B claims, 1.8M clinicians, 300K facilities – unmatched longitudinal data reach across many EHRs. 🔹 1,050 AI data scientists & engineers + $1B invested, with 50 additional healthcare engineers dedicated to hyper-personalizing care to each health system. 🔹 Care orchestration across 64 specialties + Emergency + Primary Care, infused with 60+ evidence-based guidelines. 🔹 10 business days to go live in your EHR workflows once relevant data is received – GenAI-native platform speed that no legacy system can match. 🔹 $325B in AI investments by tech giants, tapped through a portfolio of LLMs + proprietary agents to build models hyper-personalized for each system. 🔹 $57M in incremental revenue validated for just 200 PCPs – while improving care quality. ⸻ This isn’t about waiting for the next roadmap feature. It’s about giving drowning clinicians and financially strained systems outcomes at scale — now. That’s the shift RhythmX AI is driving.
12 Comments -
Jon Russell MBA
The Most Dangerous Word in Healthcare IT: “Submit Another Ticket.” It sounds harmless. Routine. Procedural. Efficient. But when clinicians hear “submit another ticket,” what they often experience is something very different: • Start over. • Not my problem. • You’re on your own. And that’s where trust begins to erode. Healthcare depends on seamless coordination. Every extra step between a caregiver and a solution compounds friction, slows care, and chips away at credibility. When tickets are closed without true resolution… When issues bounce between teams… When cross-team handoffs feel like a relay race with no finish line… The system may look efficient on a dashboard. But the experience feels fragmented. Here’s the hard truth: If the default answer to complexity is “open another ticket,” the service model needs redesign. High-performing healthcare IT teams don’t push problems downstream. They: • Own issues end-to-end • Assign single accountable leaders • Provide proactive updates — even when there’s no new information • Measure success by resolution quality, not ticket closure speed “Submit another ticket” should be rare — not reflexive. Because in healthcare, workflow friction isn’t an inconvenience. It impacts patient care, clinician morale, and organizational trust. If your service culture truly reflects Own It and Make It Easy, then every escalation should feel simpler — not heavier. Where in your organization has process replaced ownership — and what would it take to fix it? #HealthcareLeadership #HealthcareIT #OperationalExcellence #OwnershipCulture #ClinicianExperience #ExecutionMatters #LeadershipWithEdge
6 Comments -
Dr. Louise Rix
Healthcare products break the moment one of these is missing. Clinical. Product. Commercial. Most healthtech companies are built around two of the three. Clinical + Product Safe, thoughtful, well-designed. …but financially fragile. Product + Commercial Fast growth. Strong revenue. …but clinical responsibility becomes blurry. Clinical + Commercial Medically robust businesses. …but the patient experience suffers. Each pair solves one problem. And quietly creates another. Because each discipline is optimising for something different. Clinical teams protect patients. Product teams design systems. Commercial teams drive growth. Individually, those goals make perfect sense. But the moment a healthcare product scales, the forces collide. A commercial push can change who enters the system. A small change in a product flow can alter clinical decision-making. A clinical safeguard can break the experience that made the product work. Failures often don't come from one team doing something wrong. They come from the gaps between them. Healthcare products only really work when clinical safety, product design, and commercial reality are considered together from the start. Remove one, and the system eventually breaks. --- ♻️ Repost for others building safe, effective digital health. 👉 Follow Dr. Louise Rix for practical thinking on clinical product.
18 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top contentOthers named Frank Rydzewski
4 others named Frank Rydzewski are on LinkedIn
See others named Frank Rydzewski