JustReva

Most "AI for healthcare" products are just chatbots with a HIPAA badge. The actual problem in clinics isn't information retrieval. It's that the phone rings while your receptionist is checking someone in. That's it. That's the whole problem.

@jasonlk @HarryStebbings @rodriscoll The 'refused to ship it' detail is underrated. In healthcare AI we've had versions of this — systems that technically work but where the failure mode is bad enough that you don't deploy. The question isn't 'can it?' It's 'what happens when it's wrong?'

New 20VC x SaaStr out with @HarryStebbings @rodriscoll + me. Top takeaways: 1⃣Anthropic built Mythos, a model that autonomously finds zero-day vulnerabilities, including 20+ year-old bugs, and refused to ship it. Difference vs older models is "rifle vs machine gun." Cybersecurity stocks fell on the news. Backwards. Every security vendor now has to buy AI-powered vuln detection to survive. Mandatory line item in 2027 budgets. 2⃣I've rotated back to Team Sam Dario is already one of the greatest founders of all time. But I'm burned out on the endless warnings about 80% of jobs being destroyed. It's become marketing. 3⃣Trainium is now a $20B annualized business, triple-digit growth, nearly sold out Nvidia just quietly lost 10% of its potential revenue to Amazon's own silicon. 4⃣Meta's Muse Spark isn't the best model. But it proves the $14B Scale AI buy wasn't a write-off. Meta is back. And they're pivoting to closed source, which changes Llama's ecosystem. 5⃣The 60% Solution Death Spiral if your AI agent is only 60% as good as standalone, you can't charge for it. Has to be free with base product. Feature, not business. ServiceNow, HubSpot, Figma all facing this. 6⃣SpaceX filed at ~ $2T on $18.5B revenue. 108x revenue. Most expensive IPO at scale of all time "Elon discount rate" and "Elon probability of failure rate" both assumed to be zero. 6⃣Wix spent $1.6B on buybacks at $92. Stock fell 23% that week Use cash flow to buy companies and ship agents, not to buy back stock in a blip. 9⃣CIOs are token-maxing. It's another big change, when there's already been so much. Shadow AI deployments are about to get rationalized. By 2028 the default Fortune 2000 AI vendor may well be OpenAI + Microsoft because the enterprise sales motion wins, not the hottest tech.

@jasonlk The ones that survive own a workflow too painful to rip out. Not data. Not integrations. Workflow. In healthcare that's why legacy EMRs still dominate — not because they're good, but because switching costs a year of your life. The moat is friction, not features.

I could imagine a world where honestly only a few "traditional" B2B leaders survive: Salesforce ServiceNow SAP Shopify Oracle Datadog Databricks And everyone else atrophies slowly to nothingness Honestly not that hard to imagine right now, unfortunately

@gregisenberg Built this into healthcare AI. When a patient says 'I just need something for the pain' the AI can't agree — it has to know when to say 'let me transfer you.' Compliance isn't the constraint, it's the product. 'Pushes back at the right moment' is most of the engineering.

This is the problem with AI right now in ONE screenshot. LLMs are TOO agreeable. and that's actually a billion dollar opportunity. There are MULTIPLE billion dollar companies hiding in plain sight here. the entire concept is just: AI that pushes back. If AI is your employee... Well, your best employee isn't the one who says "great idea boss" to everything. it's the one who tells you your idea sucks before you waste 6 months on it. Brutally honest AI for fitness coaching. Dating profiles. Financial planning. Copywriting. Design feedback. Hiring decisions. Etc etc. The niches are endless and nobody's building for them yet. Someone please go build this

@beach_center The hiring cycle in dental never really stops. You fill one role, the next is already open. Clinics that seem to have it figured out found a way to reduce how much the front desk depends on any one person. What's been the hardest role to keep filled consistently?

Looking for your next opportunity in a positive, patient-focused dental office? We’re currently hiring: 🦷 Dental Assistant 🦷 Registered Dental Hygienist Apply on Indeed! 🌐 | vist.ly/4xz5w #JoinOurTeam #DentalCareers #NowHiring

@BeckersHR NCH building transplant services at a community system is genuinely ambitious. The part nobody discusses: how do you scale front-end ops to match clinical growth? The bottleneck is usually scheduling and intake, not surgical capacity. How are they handling that?

Listen Here: podcasts.apple.com/us/podcast/pau… In this episode, Paul Hiltz MHA FACHE, President and Chief Executive Officer of Naples Comprehensive Health - NCH, joins the podcast to discuss the introduction of transplant services at Naples Comprehensive Health and how healthcare delivery is being reshaped at his organization. He shares insights on expanding residency programs to support workforce development and meet the growing needs of patients and communities.#podcast #healthcare #leadership

@MDhelpTEK The combination is brutal — you lose the patient on the phone AND then lose revenue on the back end. Most clinic owners only find out how bad the missed call number is after someone actually runs the data. Have you seen that moment of shock when they see the number?

Busy clinic but revenue still leaking? Missed calls & unpaid claims could be the reason. Mdhelptek helps you capture every opportunity. 👉 mdhelptek.com 💬 Comment “REVENUE” #HealthcareOperations #ClinicRevenue #Mdhelptek

@drbicuspid This gap doesn't surprise me. DSOs have dedicated ops teams to evaluate tools. Private practice owners are doing that evaluation between patients on their lunch break. It's not a tech literacy gap — it's a bandwidth gap. What area are private practices most stuck on?

The #AI adoption gap between #DSOs and private #dental practices is widening, and it matters. Learn what that means for you in this webinar 👉 buff.ly/a1vKfUs

@HealthITNews Community-building in health IT is undervalued. Do you see a gap between the enterprise/health system IT community and the independent/SMB clinic side? Those operators are dealing with completely different problems and rarely have a seat at these tables.

Health IT professionals can build community with HIMSS Executive Connect healthcareitnews.com/video/health-i…

@EvanKirstel We ran into this building for healthcare clinics. The moment that surprised me: owners weren't asking 'is AI safe for my business' — they were asking 'can it answer the phone while my receptionist handles check-in.' The use case was way more basic than expected.

Trade businesses keep our world running but most still operate on paper and gut instinct. AI isn't just for tech companies — plumbers, electricians, and contractors deserve better tools too. This is an underserved market about to explode: telecomreseller.com/2026/04/02/ai-…

@kevinmd The 'design flaw' framing is right. And it's not just physicians — 35% of healthcare calls go unanswered during business hours because the whole support layer is stretched thin too. When you fix culture for physicians, you still have to fix the operational scaffolding underneath.

Physicians who graduated in the past six years stay at their first job for about two years on average. They join for financial security. They leave because of culture and leadership. That pattern is not a failure of individual physicians. It is a design flaw in how we prepare people for careers in medicine. Kara Pepper is an internal medicine physician who stayed in the same system from the end of residency until her early forties. She now coaches colleagues through career transitions, and she has mapped the process into five stages that are remarkably predictable once you see them. Stage 1 is the one nobody recognizes they are in. It is the collective shrug. Everything is hard. Health care is broken. This is just how it is. You stop imagining alternatives because the daily grind has consumed your capacity for strategic thinking. Stage 2 is the one almost everybody skips. Before you search for a new job, before you network, before you make any decision at all, you need to clear enough mental space to think strategically. Take FMLA if you have it. Drop the committee. Cut the obligations that are consuming energy without returning value. The goal is to get your nervous system out of chronic stress long enough to think with clarity instead of desperation. Kara's observation here is sharp: if you make career decisions from a depleted state, you will recreate the same dynamic in the new place. The geography changes. The pattern does not. Stage 3 is discovery, and it starts with a counterintuitive question: what makes you angry? Not mildly frustrated. Genuinely resentful. Kara describes getting furious when friends asked her to play tennis at 10 in the morning. She did not want to play tennis. She wanted autonomy over her schedule. That resentment was pointing directly at what she needed, and she had been ignoring the signal for years. Stage 4 is community. Medicine trains you to solve problems alone, but career reinvention is not a clinical problem. Go to conferences outside medicine. Look at what physicians are building in startups, consulting, tech, ancillary services. The options are not "stay in clinical medicine" or "leave medicine entirely." There are hundreds of paths that use your skills in ways your training never showed you. Stage 5 is tiny action. This is where people stall because they confuse "action" with "demolition." Kara's own exit was messy and unplanned, and she will tell you directly: do not do it that way. But doing nothing is not caution. It is the same trap wearing a different outfit. One financial calculation. One mentor meeting. One job search. Pick the smallest possible move that makes the next move visible. Worth saving if you are coaching a colleague through this process or if you are somewhere in these stages yourself and could not name where until now. Search "The Podcast by KevinMD" wherever you listen to podcasts. Link in the replies. Which of these five stages did you get stuck in the longest? #ThePodcastbyKevinMD #PhysicianCareerChange

@FierceHealth The specialty spread is the interesting part. Which ones saw the biggest delta — and did the data break out whether employed vs. independent practice compensation moved differently? Independent clinics are dealing with cost pressure from every direction right now.

Physician compensation up 3% in 2025, but not all specialties saw raises: Medscape bit.ly/4tKz4FQ

@EvanKirstel Mostly agree — but the 'real conversations' bar is higher than most vendors admit. In healthcare, routing is 80% of the job. The other 20% is knowing when NOT to handle it and escalating cleanly. Most enterprise voice AI fails at exactly that edge case.

Pressing zero to reach a human should not be the default customer experience in 2026. Enterprise voice AI is finally getting good enough to handle real conversations — not just route calls. Here's what actually works: telecomreseller.com/2026/04/01/ent…

@jaltma The Whole Foods moment is underrated as a case study. Most companies treat existential pressure as a reason to pivot — Instacart treated it as a reason to go deeper. What's the forcing function that actually makes you stop debating and just commit?

Never let a good crisis go to waste. One of the scariest moment's in Instacart's history was when Amazon bought Whole Foods...their biggest competitor buying their biggest client. Max tells the story of how they locked in (before locking in was a term) and made it an advantage.

@patrickc @huggingface @Cloudflare @OpenRouter @firecrawl @flydotio @Amplitude_HQ @mixpanel @inngest The stack keeps getting richer. Curious how you're thinking about regulated industries — healthcare and finance devs still have to navigate a layer of compliance before they can actually ship. Does Projects have a path to making that less painful?

Just launched some great new providers in Stripe Projects (projects.dev): @huggingface @Cloudflare @OpenRouter @firecrawl @flydotio @Amplitude_HQ @mixpanel @inngest Many more coming later this month.

@mpappas74 @phonely_ai Most clinics don't know their actual missed call rate — they assume it's fine because the receptionist is "always there." But 11am–1pm and 4–6pm are brutal. Phone rings most, staff busiest with check-in and checkout. What kind of clinic are you looking at?

@IamJustReva @phonely_ai curious to know more about this

We raised a $16M Series A to fix the biggest blind spot in business: phone calls. $2 billion is transacted over the phone every day. And most companies have zero insight into what's happening on those calls. Why? There's no analytics. No optimization. No way to know what's working and what's losing you money. So we built the optimization engine from scratch. Voice agents that sound human. A/B testing for phone calls. The most advanced voice AI analytics on the market. One customer generated $10M in net new revenue in three months. We've gone from hundreds of calls to millions. Outperforming the best human reps. Getting better every day. Every business leader keeps getting asked: how will you use AI? Now you have an answer. Start with the phone. Optimize your calls with @phonely_ai . -- Excited to partner with @Base10Partners, @ycombinator, and more to build the future of voice.

First time a clinic owner saw the missed call data, she went quiet for a second. Then: "We thought we were doing fine." They were missing 38% of calls during the 11am–1pm window. No one had ever shown them the number before. That moment is why I'm building this.

The vehicle causality distinction actually matters most when the AI is wrong. A clinician can tolerate an AI that sometimes misses — what they can't tolerate is one that confidently acts as if it understands. The error mode changes everything. Does your work focus more on that failure transparency layer?

@IamJustReva @AlexLerchner @PhilPapers_CDP You make a valid point, thanks you.

The Abstraction Fallacy: Why AI Can Simulate But Not Instantiate Consciousness philpapers.org/rec/LERTAF ✍️ @AlexLerchner via @PhilPapers_CDP 👉 Synopsis: "The framework proposed here explicitly separates simulation (behavioral mimicry driven by vehicle causality) from instantiation (intrinsic physical constitution driven by content causality). Establishing this ontological boundary shows why algorithmic symbol manipulation is structurally incapable of instantiating experience. Ultimately, this framework offers a physically grounded refutation of computational functionalism to resolve the current uncertainty surrounding AI consciousness" @Corix_JC @ahier @maponi @CEO_AISOMA @dinisguarda @JagersbergKnut @RamonaEid @FernandaKellner @PVynckier @EstelaMandela @sonu_monika @ChuckDBrooks @SusanHayes_ @theomitsa @TarakRindani @FrRonconi @Nicochan33 @JoannMoretti @NeiraOsci @Khulood_Almani @Fabriziobustama @TheAIObserverX @pchamard @mikeflache @TysonLester @Eli_Krumova @smaksked @amalmerzouk @Analytics_699 @sminaev2015 @WillyRayNick @RLDI_Lamy

Healthcare is where it hits hardest — a missed call isn't just lost revenue, it's a patient who calls a competitor or just doesn't book at all. Most clinic owners have no idea what their actual miss rate looks like during lunch hour or end of day. What kind of practices are you thinking about?

She Googled "dentist near me." You were on page 2. She booked with page 1. 75% of patients never scroll past the first page. Where do you rank? Free scan → justreva.com/grader