“The builder’s hat must be on; this is the era for everybody to try to build, try to think, and try to dream of what’s possible.”
— David Friedberg
OpenCare
I’ve always heard that if you’re going to build an app, build something that solves a problem you’re having or that isn’t solved yet. OpenCare is helping me care for my aging parents, and it’s just one of the many problems the Poor People App is solving.
Hard-E
Hard-E is a digital employee. Connected to a company CRM, it can email, text, and learn. Special knowledge files, the ability to synthesize material costs, provide measurement and pricing data. Available hands-free in voice mode.
All Angles
A five-step interactive journey map that replaces the cold contact form. Vision models read the house. Perplexity researches the property. Sales gets a briefed lead. The homeowner feels in control the whole time — which, by design, they are.
Ethos
Ethos is my first autonomous agent — built on Hermes. It coordinates care for my parents through a live OpenCare API, runs a video intelligence pipeline that turns YouTube into wiki-linked investment signals and more. Employee, volunteer, partner.
What's Running Right Now
LIVE
Poor People App / OpenCare — When you have a problem, build a solution
I was drowning in group texts trying to coordinate my parents' care: doctors, insurance
renewals, medications, siblings who needed updates but didn't need another app. So I
built OpenCare:
a care coordination platform where circles of helpers organize around the people who
matter, with tasks, reminders, and daily digests that reach even the family members
who barely check email.
Then the mission grew. OpenRecovery
applies the same circle model to substance use and dual diagnosis recovery — 10
modalities, daily check-ins, journaling, and Recover E, an AI voice companion
that's sharp, challenging, and built for people who want to do the work, not sit back
and wait. It's not clinical. It's not cookie-cutter.
The whole thing lives under the Poor People App — a platform for underserved
communities where the AI handles the 95% of bureaucratic work that's purely
procedural, and human Sentinels handle the 5% that requires real judgment. Identity
on Nostr. Care that's free. Infrastructure for the people who can't afford to wait
for the system to work.
53M unpaid US family caregivers · ~$30/mo server cost · Running on one EC2 instance.
opencare.poorpeople.app
·
poorpeople.app
PRODUCTION
Hard-E — AI Sales Director for Contractors
I built Hard-E to prove that AI agents can do real work inside a business — not
answer generic questions, but actually operate a CRM, hold a phone conversation,
and get smarter over time.
Hard-E is a full production agent for home service contractors. It connects directly
to a CRM with 21 tools that cover the entire sales workflow: customer lookup, job
management, pipeline stage updates, SMS, email, appointment scheduling, proposals,
financials, pricing, and document handling. Some of those CRM endpoints don’t
even exist in the official API — I reverse-engineered them from browser dev
tools to get features like email access and note notifications working.
The voice mode is what sets it apart. Deepgram transcribes speech, Claude Sonnet 4
reasons and picks which tools to call, and Cartesia speaks the response back —
all in real time over WebRTC. A user can ask Hard-E to check on a customer’s
job status mid-conversation and get the answer without the call breaking stride.
The entire voice pipeline runs as a separate Pipecat process, and I built a custom
WebGL shader orb that visualizes the conversation in real time.
It remembers. A background service watches for idle sessions, extracts what was
learned — customer preferences, pricing patterns, workflow corrections —
and stores it. Next conversation, that knowledge is already in context. Voice
sessions get the same treatment: when a call ends, the system summarizes and saves
anything worth keeping.
Multi-tenant, encrypted credentials, tenant-isolated knowledge bases, invite-based
trials. Built to clone — swap the CRM plugin, rewrite the personality file,
upload new training documents, and it becomes a sales director for a different
industry. I’ve already started doing that.
FastAPI · React · Redis · S3 · Claude Sonnet 4 · Pipecat · Deepgram · Cartesia · EC2
harde.app
TECHNICAL LEAD
All Angles — Remodeling How Remodeling Gets Sold
Most contractor websites are brochures with a contact form bolted on. This one had to
be different — the founder spent 30 years watching the industry get it wrong. Hard
sells, vague estimates, leads that go nowhere. All Angles launched as a startup built
on the opposite: educate first, earn trust, then ask permission.
So I built the entire conversion around that. No contact form. Instead, a guided
interactive journey that walks homeowners through the process at their own pace. By
the time they reach the end, they've learned something, they've uploaded photos of
their home, and they've decided on their own terms that they're ready. That's when
the system asks for one thing: permission to come measure.
Behind the curtain, uploaded photos hit S3 and GPT-4o Vision reads the exterior —
siding material, trim condition, visible damage. Perplexity researches the property.
The contractor receives an AI-briefed dossier on a lead who already said yes, before
he's picked up the phone.
Next.js 14 · Supabase · GPT-4o Vision · Perplexity · AWS (SES, S3, DynamoDB) · James Hardie Elite Preferred
aaexterior.com
ALWAYS ON
Ethos — Personal AI, Running Continuously
Ethos runs 24/7 on AWS Lightsail — voice and text over Telegram. It coordinates care
tasks via the OpenCare API, processes video research through a multi-lens pipeline,
and maintains a living knowledge base across investing, technology, and frontier
science. Over 100 pages, growing weekly. It delegates to cheaper models when it can,
resets its own sessions, curates its own skills. It's not a product. It's mine.
$2–4/day · 100+ wiki pages · Hermes Agent · Voice-enabled
Internal system
