Unknown Behavior Independent Practice
Founder of ZONA. I enjoy building consumer products and architecting the revenue systems underneath them. Most recently, I've shipped applied AI products into real operations, in the US and abroad, to help businesses grow.
Selected work
The product, and the business underneath it.
The flagship. A real-time Miami nightlife platform I started in 2023 and launched in 2024. I led the team that shipped it across iOS and Android and took it to 150K+ downloads and 25+ venue partners, then built the business underneath: a partner portal, an ads product, and a lifecycle engine. Owned end to end, product to revenue.
Forward-deployed in a Lima clinic: I embedded in their operations, mapped the workflow and decision-making, then designed and rolled out a pre-arrival digital intake system, winning buy-in over two competing options. I trained the teams and ran a 30-day post-launch KPI cadence with leadership. Wait time down 50% (20 to 10 minutes), no-shows down ~40%, no added staff.
An NYU Berlin team project I led. We took a vague idea and shaped it into a focused product: runtime governance for AI in medication. We presented to Google and Join Capital and won the final pitch before a faculty panel. Sentra is middleware between an AI prescription system and the pharmacy: it validates every AI-generated medication order against live patient data and returns allow, block, or escalate in under 100 milliseconds, with an audit trail built to the EU AI Act. AI safety, applied where being wrong is not an option.
A voice agent you talk to out loud that answers in real time, running locally. I built the full pipeline, streaming speech-to-text, a Claude brain with live tool-use, and streaming text-to-speech, on Pipecat with Deepgram, Cartesia, and Silero VAD, then made it smooth on real hardware: solving audio sample-rate sync, acoustic echo, Bluetooth device routing, and the dependency pinning that real-time voice demands. Configurable personality, shipped as an open template anyone can make their own.
A value-based-care problem taken to a live, working product. It ranks open care gaps by clinical urgency, response likelihood, and equity, then drafts personalized patient outreach with Claude (using prompt caching), behind a public methodology card. Applied AI where the prioritization has to be defensible, not a black box.
Two AI models check each other's work so you don't have to: a worker writes, an overseer stress-tests it, and agents respawn with memory after repeated failures. A practical take on AI reliability, catching the bad output before it ships.
A photo of the affected area plus a symptom description in, schema-valid pre-visit triage JSON out: ranked causes, red flags, and questions for the doctor, built on Claude's vision API. It ships with a side-by-side eval harness comparing image-plus-text against text-only, because in clinical AI you prove the approach is better, you don't assume it. (v1 fine-tuned an open model with LoRA; v2 is the multi-modal build.)
A multi-vendor integration was quietly leaking bad data. I found the breaks, built validation at every join, and put it all in one view leadership could actually read. Mismatches down 25%, manual reporting down 40%, breaks caught at ingest instead of weeks later.
Surfaces
The products and platforms, consumer and B2B.
The ZONA app: real-time nightlife, where tonight is actually happening, across iOS and Android.
zona.miami: 500+ programmatic venue and event pages indexed on Google and Bing, turning Miami nightlife searches into organic discovery.
The getzona.app header CTA, treated as a growth lever. A device-aware Download button that routes iPhone to the App Store and Android to Google Play, consistent across all 10 pages, with every click instrumented in GA4 and an automated weekly click-through-rate report. Pointed at installs and measured, not guessed.
The supply-side data layer the decision engine runs on. I built the pipeline that keeps 400+ venues accurate and current, reconciled against authoritative sources like Google Places, because 'where is open tonight' is only as good as the data beneath it.
ZONA's retention runs on push, so I built the nightly decision engine that picks who gets messaged and when, under hard guardrails like frequency caps, recency floors, and open-rate backoff, and monitors its own tap rates to pause weak triggers. Lifecycle, not blasts.
A live ranking of every venue by real-time demand, using an exponential time-decay model so current momentum outweighs history. It shows each venue its exact rank, drives competition, and feeds the ads funnel.
On The List Miami: ZONA's nightlife briefing, a direct channel I own that turns the app's audience into recurring attention. Editorial takes, list rankings, and venue intel in the inbox every month.
The B2B side. Venues run their presence, events, and tables in one place. I built it and onboarded 25+ of them.
The supply side that fills the rooms. Promoters track their links, guest lists, and payouts in one place.
Audience, turned into revenue. An ad product that puts venues and brands in front of the right crowd at the right moment.
Features
The systems that make it work.
Contact
For roles, collaborations, and product work.