Presenter view
End with sequencing and restraint: what ships first, how the current product migrates and which attractive ideas should remain out of scope.
Opening cue
01 · Delivery horizons
Two weeks
A clinician can review a safe weekly queue generated from synthetic/fixture panels.
Scheduled Temporal parent workflow
SQL/change pre-filter + bounded top-candidate fan-out
Reuse existing single-patient reasoning core
Evidence-backed ranked queue; human approval only
Audit trail, budget ceiling and engineered fixture panel
Two months
Panel and inbox workflows share durable state, event triggers and deterministic evaluation.
PatientStateChanged invalidation + freshness policies
Inbox classify/draft/hold/escalate workflow
Typed agent/tool contracts + improved grounding reviewer
Feedback and explicit preference/constraint records
Replay, shadow, canary and cost/latency dashboards
Two quarters
Three workflow shapes run on one governed, measurable backplane.
Practice-improvement analytics and patient drill-down
Calibrated prioritization by organization and population
Selective low-risk operational autonomy
Outcome-linked memory with review and expiry
Broader connectors, policy layering and continuous evaluation
02 · Shared backplane migration
Define shared contracts
PatientState, ContextPacket, DecisionCandidate, SpecialistRequest, ActionRequest and evaluation events.
Extract reusable core
Context Builder, Tool Gateway, Agent Directory, reconciler, grounding and action policy become host-agnostic modules.
Add two runtimes
Current synchronous session runtime and Temporal async runtime call the same versioned core.
Shadow + compare
Run the new path beside the current batch path; compare actions, citations, cost and failures.
Migrate one workflow at a time
Panel first, then inbox, then practice insights; no rewrite and no team freeze.
03 · Deliberate debt
Hand-tuned ranking weights
Fast, explainable starting point
Calibrate from adjudicated panels; retain deterministic guardrails
PostgreSQL + pgvector
Simpler security and transactional provenance
Split only when measured recall/latency/scale thresholds fail
Limited condition roster
Ship safe depth for common chronic disease
Add agents only with owner, tools, eval set and lifecycle plan
Human approval for every clinical action
Controls blast radius while quality is learned
Expand only narrow, reversible operational autonomy
Partial event coverage
Start with labs, medications, messages and scheduled review
Add claims/transitions after source-specific quality contracts
The strong answer is not “this is temporary.” It is “this is the metric or failure mode that will force us to replace it.”
04 · Non-goals
Do not replace the EHR or authoritative source systems.
Do not run frontier reasoning across every patient and condition.
Do not create a separate “Praxis Async” reasoning stack.
Do not learn clinical truth from rejection behavior alone.
Do not auto-send clinical messages or execute high-risk actions.
Do not build a generic agent marketplace or recursive agent hierarchy.
Do not claim clinical thresholds without clinician ownership and evaluation.
Do not introduce a dedicated vector database before pgvector proves insufficient.
05 · Assumptions + closing position
PostgreSQL, object storage, Temporal, hosted observability and frontier APIs already exist.
EHR write-back supports only a narrow subset of action types.
Every clinical action has a named accountable human.
No PHI reaches a vendor without a BAA; logs and traces minimize PHI.
Clinician attention—not compute—is the scarcest resource.
Critical clinical thresholds, labels and autonomy policies require clinician ownership.
All patients and examples used in development and evaluation are synthetic or approved fixtures.
Final recommendation
“Ship a safe, explainable weekly panel queue first. Use what it teaches about state, cost, trust and evaluation to earn inbox automation and practice-level learning.”
AI-use disclosure cue
“I used AI to organize the artifact, challenge gaps and accelerate presentation implementation. I overrode it on scope, the explicit assumptions, workflow boundaries, the decision to separate memory types, and the requirement that every value remain evaluation-tunable. I remain accountable for the architecture.”