THE NEXT PATIENTAI-driven health care intel

Technology architecture · tools serve the workflow

The technology stack

The stack is organized by function: acquire evidence, do the work, preserve state, store assets, deliver the product, and observe what actually happened. Any individual AI model can be replaced without losing the business.

Models think. Systems remember. Live surfaces prove.
The redesign

GitHub is the control record, not the binary bucket. R2 is the working binary store. Beehiiv is the publication engine. The browser is an operating surface—not a durable integration.

End-to-end architecture

From outside signal to observed reader

The blue return line matters as much as the green production line. Without received-form evidence, the system knows what it sent—not what the reader saw.

Production flowObservation and learningHuman or promotion gateOpen integration
The Next Patient layered technology pipeline Sources flow through AI and automation into durable GitHub state and R2 assets, then into Beehiiv and Cloudflare surfaces. Gmail, analytics, replies, and errors return observed evidence. 01 · ACQUIRE Evidence inputs Primary records Gmail + newsletters Grok / web research APIs + feeds SOURCE + DATE REQUIRED 02 · ORCHESTRATE Work layer ChatGPT / Codex Browser operation APIs / MCP connectors Scheduled workflows STEVE GATES RELEASE 03 · REMEMBER Durable state GitHub master NOW + protocol Manifests + versions Obsidian local view CANONICAL TEXT RECORD 04 · STORE Binary assets Cloudflare R2 Images + proofs Generated working files Drive static archive POINTERS LIVE IN GITHUB 05 · DELIVER Product surfaces Beehiiv email + site Cloudflare ops + read Beehiiv checkout Gmail + Telegram SENDER GATE OPEN 06 · OBSERVE Received email · render · links · delivery · clicks · replies · failures Evidence returns to the work layer and the durable record deliberately—not as inferred rules. MIGRATION LANE · Intel Filter · Reading Room · recurring health routines
Preferred order: connector or API for repeated work; browser for configuration and visual proof; manual action only at the real gate.The dotted migration lane is acknowledged but not designed here.

Management view

What each layer owns

A tool is not an owner. Ownership sits with the layer's contract, state, and proof requirement.

LayerPrimary systemContractFailure test
AcquisitionGrok, Gmail, web, APIsEvidence package with source, date, relevance, and uncertainty.Can another worker verify the claim without the chat?
OrchestrationChatGPT / CodexNamed job, named inputs, inspectable output, and explicit gate.Could the work resume tomorrow from files alone?
Durable stateGitHub master + Obsidian viewCurrent routing, versions, protocol, and provenance.Does NOW.md point to one current artifact?
Binary storageCloudflare R2Stable object and manifest pointer. Drive remains a static archive.Can the named image or proof be retrieved directly?
DeliveryBeehiiv + CloudflareEmail, site, ops, and reading surfaces as the reader sees them.Did the delivered product arrive, render, and link correctly?
ObservationGmail, analytics, replies, logsReal-world evidence tied back to the named run.Are corrections based on observation rather than model confidence?

Build now

  • Prove Beehiiv sender and received-form QA.
  • Keep GitHub and R2 roles sharply separated.
  • Preserve one named run package per issue.
  • Return live evidence to the next run.

Do not optimize yet

  • Multi-agent autonomy before the line repeats.
  • Automated editorial promotion.
  • A new platform to replace the current spine.
  • The legacy routines until the issue factory is stable.