Bogdan Predusca

If you’re building a product, start with people — not features. Too often, founders get lost polishing details, designing flows, and adding “one more improvement,” only to realize users don’t really need it. At WeFi, we do it the other way around. We listen first. Every new idea starts with conversations — what the community wants, what problems they actually face. If people say, “Yes, that’s exactly what I need,” — we build it. If not, we move on. Simple as that. It’s not just a community-first approach — it’s how we save time, focus, and energy for what really matters. Because in a startup, the biggest risk isn’t being wrong — it’s building something nobody asked for.

26

2 Comments

80% of health systems are racing to plug AI into revenue cycles. The quiet 20%? They might be spotting the hidden costs surfacing right now. 🔍 My take: This surge from 58% to 80% adoption in just two years shows healthcare's desperation to fix broken revenue cycles. But the real story? Those implementation gaps are massive. For those of us building AI solutions in healthcare, this trend reveals a critical disconnect: 🏥 Large systems (64% piloting) vs smaller hospitals (20% piloting) 💰 $80M revenue at risk for billion-dollar organizations ⚡ Methodist Health System saved 5,559 hours in 8 months 🔧 Integration challenges still blocking 60% from full deployment The math is brutal. 89% say coding errors hurt revenue. That's not a tech problem - it's a survival problem. What excites me most? AI isn't just automating tasks. It's uncovering missed reimbursements and filling documentation gaps that human eyes miss. Lastly, I think it's important to keep an eye on the smaller health systems getting left behind. They need AI solutions designed for their scale and budget constraints. The real winners won't just implement AI. They'll solve the integration puzzle that's keeping most organizations stuck in pilot purgatory. What's your experience with AI in healthcare revenue management? Are you seeing similar adoption patterns? #HealthcareAI #RevenueManagement #AIImplementation 𝐒𝐨𝐮𝐫𝐜𝐞: https://lnkd.in/gF5-n9Mu

9

🚀 8 founders said “yes” before the product was ready. Before Launchyfi had a dashboard, a pricing page, or even a public demo, 8 founders agreed to become our design partners. That’s not luck. That’s validation the hard way. We didn’t start by building features. We started by sitting down with people who feel the pain: founders and GTM leaders buried under tools, duct-taped workflows, and spreadsheets. Every single one said the same thing in different words: “We don’t need another tool. We need a system that actually thinks.” That’s when we knew we were onto something. In the past, we’ve made the classic mistake: Building too long in a vacuum. Perfecting what no one had validated yet. This time, we flipped the script, and it changed everything. Working with founders from day one helped us move faster, build smarter, and cut months of wasted development. Every feedback call, every live teardown, every workflow review became a blueprint for Launchyfi’s core system. The AI GTM Engineer that plans, builds and runs your GTM for you. It hasn’t been smooth. 😱 We’ve scrapped prototypes. 🔨 We’ve changed workflows we thought were perfect. 🧠 But that’s what happens when you build with your customers, not for them. But here’s what this stage has taught us: ✨ Grit > hype. 🤝 Collaboration > assumptions. 🔁 Validation > velocity. These 8 founders didn’t just buy into a product. They bought into us, and the mission to engineer growth that actually compounds. We didn’t pitch another tool to add to the 16,000 others already crowding GTM stacks. We showed them a different path. A real solution. A GTM Engineer built to be 10× faster, 10× smarter, and 10× more cost-efficient than the fragmented systems most teams are stuck with today. We’re now working with them to refine, validate, and battle-test the platform ahead of public release. Launchyfi was never meant to be built in isolation. It’s being engineered in the wild with the operators who need it most. Big thank you to our design partners for the trust, time, and brutal feedback. You’re shaping something far bigger than another GTM tool. 👀 Public unveiling soon. Want to see it before everyone else? Come see us at Web Summit #Lisbon For now here is a sneak peak... #GTM #Startups #SaaS #AI #Launchyfi #FounderJourney #BuildInPublic

16

8 Comments

Unpopular opinion but defining PMF is the biggest mistake pre-PMF companies make. The moment you think "this is what PMF means for us," you create a false finish line. You start bending reality to match your definition rather than letting the market tell you what matters. PMF isn't a metric you define upfront. It's a state you recognise in retrospect. You know you have it when customers pull the product from you, when they get angry if it goes down, when they recommend it without being asked. But if you define PMF as "X metric," you'll optimize for the metric instead of the underlying truth it's supposed to represent. Stay radically open to redefining the problem itself, not just iterating on solutions to a problem you locked in too early.

18

5 Comments

Most startup ideas fail because founders place the wrong bet. Not too early. Not too late. Just… in the wrong direction. Andreessen Horowitz’s 2026 roadmap is really a map of where the next winning bets are forming: https://lnkd.in/gx3MEUjr If you’re choosing an idea this year, think in bets, not features 👇 𝗕𝗲𝘁 𝟭: 𝗜𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲 𝗕𝗲𝘁𝘀 (Slow to build. Hard to replace.) • Cleaning & governing unstructured enterprise data • Infra that survives agent-scale workloads • AI-native data stacks built for agents • Software designed for machine readability If your idea only works for humans → it’s fragile. 𝗕𝗲𝘁 𝟮: 𝗔𝗴𝗲𝗻𝘁-𝗙𝗶𝗿𝘀𝘁 𝗕𝗲𝘁𝘀 (Software stops waiting for clicks.) • AI agents replacing traditional UIs • Voice agents managing full workflows • Prompt-free, proactive applications • Multi-agent coordination platforms The best ideas won’t be used. They’ll be delegated to. 𝗕𝗲𝘁 𝟯: 𝗩𝗲𝗿𝘁𝗶𝗰𝗮𝗹 𝗔𝗜 𝗕𝗲𝘁𝘀 (Where AI prints real ROI.) • AI for manufacturing, logistics, agriculture • Systems coordinating multiple stakeholders • Outcome-based pricing replacing seat licenses • Software sold at company formation Unsexy industry. Massive upside. 𝗕𝗲𝘁 𝟰: 𝗠𝗲𝗱𝗶𝗮 & 𝗖𝗿𝗲𝗮𝘁𝗶𝘃𝗶𝘁𝘆 𝗕𝗲𝘁𝘀 (Content becomes interactive infrastructure.) • Multimodal creation (text, audio, video) • Stateful, interactive video • World models & explorable environments • Personalization to the individual Static content doesn’t scale anymore. 𝗕𝗲𝘁 𝟱: 𝗣𝗵𝘆𝘀𝗶𝗰𝗮𝗹 𝗪𝗼𝗿𝗹𝗱 𝗕𝗲𝘁𝘀 (Atoms finally meet software.) • AI-native industrial companies • Software-controlled machines & energy • Real-time observability of cities • Autonomous scientific labs The next trillion-dollar companies will touch the physical world. 𝗕𝗲𝘁 𝟲: 𝗙𝗶𝗻𝗮𝗻𝗰𝗶𝗮𝗹 & 𝗧𝗿𝘂𝘀𝘁 𝗕𝗲𝘁𝘀 (Rails > apps.) • Stablecoins as bank-ledger upgrades • AI-native banking & insurance infra • Onchain real-world assets • Identity for autonomous agents Money is being rebuilt from scratch. 𝗧𝗵𝗲 𝗙𝗼𝘂𝗻𝗱𝗲𝗿 𝗥𝗲𝗮𝗹𝗶𝘁𝘆 Every startup starts with one bet. The great ones expand into adjacent bets. The real question isn’t: “Is this a good idea?” It’s: 𝗜𝘀 𝘁𝗵𝗶𝘀 𝘁𝗵𝗲 𝗯𝗲𝘁 𝗜 𝘄𝗮𝗻𝘁 𝘁𝗼 𝗰𝗼𝗺𝗽𝗼𝘂𝗻𝗱 𝗳𝗼𝗿 𝘁𝗵𝗲 𝗻𝗲𝘅𝘁 𝟭𝟬 𝘆𝗲𝗮𝗿𝘀?” ♻️ Repost if this reframed how you think about startup ideas

9

I've looked at a lot of YC batches. The bio and health lineup in P26 is STACKED. https://lnkd.in/gQKxpe9t Adialante - Compact, mobile whole-body MRI to make cancer screening accessible at hundreds per scan FinalDose - A programmable drug that reads DNA inside the cell and destroys it if it's diseased, starting with previously undruggable cancer targets Voquill (YC P26) - An AI coworker for pathologists that learns your reporting style and produces sign-out ready reports in real time Clara - An AI primary care doctor with a licensed clinician reviewing every medical decision Lumius (YC P26) - Affordable, real-time 3D ultrasound, starting with vascular access

112

7 Comments