Live tools, the Shepherd loop we run in clinic, and sketches that are still sketches. No invented case-study percentages.
Record the visit, transcribe into client-friendly language, push into Shepherd with one click. Built for Hearthstone because nothing on the market did the whole loop.
Problem. Vendors sell a transcriber or a discharge PDF. The evening is still spent retyping into the PMS.
Approach. Custom integration into Shepherd: audio → owner language → medical record. Doctor still reads and signs.
Status, honestly. In production at our clinic. Not a public app you can log into. The method is in the building-tools essay.
Patient drop-off take-homes: why they are here, what we are doing, pickup and overnight watch-fors in owner language.
Problem. Drop-off paperwork is rushed, handwritten, and generates the 8pm callback.
Approach. Templates for the day-admit visit plus an AI fill-in, reviewed before pickup.
Status, honestly. Live at dischargepro.lovable.app. A draft. The doctor still edits.
Post-op take-home: incision, cone, pain meds, what redness is, what swelling is not.
Problem. Surgery aftercare is a different sheet than a drop-off. One form printed the wrong paragraph.
Approach. A separate app at discharge.peltonsites.com. Same rule: I wrote the templates, software fills the visit, I read it.
Status, honestly. Live. Not a diagnosis. Not DischargePro — different job.
Waiting-room pet photo slideshow. Built because the consumer options were junk.
Problem. Clinics want patient photos on a loop. Chromecast folders and 2009 dental vendors are not a product.
Approach. A dedicated gallery at pawfect-gallery.lovable.app for the lobby screen.
Status, honestly. Live. It shows pets, not a diagnosis.
First-pass lab results in English an owner can follow. The doctor still interprets the case.
Problem. Lab PDFs are a wall of abbreviations. The callback becomes a translation job.
Approach. A small interpreter at antech-interp.lovable.app. Draft explanation, human review.
Status, honestly. A draft. Not a diagnosis. Open it, then still read the actual panel.
Swipeable Bitcoin cards with optional sign-in so progress sticks.
Problem. Most Bitcoin explainers are either hype or a whitepaper.
Approach. Short cards, written for busy people, sitting next to the longer guides.
Status, honestly. Card volume depends on what has been approved in the admin tools. The written guides are always there.
A simple model of recurring buys for a personal or practice reserve.
Problem. People skip the arithmetic and jump to a price target.
Approach. Show total invested, coins accumulated, and how duration changes the picture. Not a forecast.
Status, honestly. Uses a price you set. It does not predict returns.
How we keep master templates and let software fill the visit-specific pieces.
Problem. Every doctor writes a slightly different 'bland diet' paragraph.
Approach. Stable templates, AI polish, human review. Documented in the AI Lab.
Status, honestly. A workflow, not a SaaS listing. Read the guide; steal the structure.
A concept for collecting history before the appointment and handing the doctor a clean summary.
Problem. Intake misses timeline, meds, or red flags, and the appointment spends its first ten minutes catching up.
Approach. Guided questions, red-flag routing, RAG over our own protocols — not WebMD.
Status, honestly. Not a shipped product. The article is the design note.
Lunch-and-learn material so the front desk is not blindsided if a client asks about bitcoin payments or scams.
Problem. A clinic cannot offer, or even discuss, Bitcoin if only the owner has read a thread.
Approach. Short sessions: wallets, seed phrases, 'nobody legitimate asks for your 12 words.'
Status, honestly. Curriculum, not a certified course. Start with the public guides on this site.
Ideas, not roadmaps. If one of these would help your practice, say so on the contact page.
Educational only. Not medical, financial, or legal advice.