Session Prep Pro

Session Prep Pro is a full-stack clinical productivity tool built for licensed therapists and counselors in private practice. Clinicians maintain a private roster of clients (using anonymized identifiers such as initials or pseudonyms), each with clinician-authored background context. Before a session, the clinician pastes in raw notes — progress notes, EHR exports, prior session summaries — and the system uses a large language model (GPT-4o via the OpenAI API) to generate a structured Pre-Session Brief with four sections:

• Sensitivities — clinically sensitive topics or dynamics extracted from the notes, each tagged with an urgency level (high / medium / low)
• Key Themes — recurring therapeutic themes across session history
• Follow-ups — specific items the client raised that need to be checked on next session
• Opening Recommendations — concrete, actionable language for how to open the upcoming session

Brief generation runs asynchronously: the server returns immediately, processes the AI call in the background, and the frontend polls every 2 seconds until the brief is complete. A dashboard gives the clinician an at-a-glance view of client count, briefs generated, and recent activity.

This is a clinician-side workspace only — it is not a therapy tool or patient-facing product. It is purely internal, scoped to each individual therapist's own clients, and designed entirely around the pre-session preparation workflow.

A working prototype already exists (built on Replit.ai) demonstrating the core flow end-to-end, including a monorepo architecture (pnpm workspaces), contract-first API design (OpenAPI + codegen), Clerk-based multi-tenant authentication, and a React/TypeScript frontend. This gives the capstone team a running start, with the bulk of the work centered on hardening, extending, and validating the system for a real pilot.

Clinicians spend significant time before each session manually rereading their own notes to reconstruct context: what was discussed last time, what concerns were flagged, and what needs follow-up. This "pre-session cognitive load" is a well-documented burden in clinical workflows, and it competes directly with the clinician's limited prep time and attention. Left unaddressed, it can contribute to missed follow-ups, disjointed session openings, and — over time — ruptures in the therapeutic relationship that reduce the effectiveness of treatment.

Session Prep Pro addresses this by turning a clinician's own raw notes into a compact, clinically organized brief that surfaces the most actionable information right before a session starts, without requiring the clinician to re-read an entire history of notes each time.

The primary objective is to take the existing prototype and develop it into a robust, pilot-ready application that can be tested with counseling and counselor education professionals in the faculty sponsor's network.

Core Objectives

• Harden and extend the existing prototype (client roster, brief generation, dashboard) into a stable, well-tested application suitable for real clinicians to use
• Review and improve the security and multi-tenancy model so that each clinician's data is properly isolated and protected
• Design the pilot study itself: onboarding flow, feedback collection from pilot clinicians, and a basic evaluation of whether briefs are perceived as accurate and useful
• Produce documentation (architecture, data model, setup/deployment) sufficient for a future team or the faculty sponsor to maintain the system after the capstone ends

Stretch Goals

• Replace the inline async job handling with a proper background job queue (e.g., BullMQ + Redis) for reliability at scale
• PDF export of a generated brief for the clinician's own records
• Brief versioning — track multiple brief generations per session date
• Customizable AI prompts, letting a clinician emphasize a particular therapeutic framing (e.g., CBT vs. psychodynamic)
• An audit log of who accessed what and when, aligned with HIPAA-style access logging practices
• A mobile-responsive or native-mobile (e.g., Expo-wrapped) experience for offline access

• Solid experience with a modern web frontend framework (e.g., React) and TypeScript
• Experience building or consuming REST APIs, including request/response design and basic authentication concepts
• Working knowledge of relational databases and SQL (schema design, queries, migrations)
• Comfort reading and adapting to an existing, non-trivial codebase rather than starting from scratch
• Basic familiarity with calling a third-party LLM API (e.g., OpenAI, Anthropic) and handling asynchronous responses
• Ability to reason carefully about data privacy and access control, given the sensitivity of the data involved

• Experience with TanStack Query (or a similar server-state library) and patterns like cache invalidation and polling
• Experience with schema-driven form validation (e.g., Zod, react-hook-form)
• Familiarity with contract-first / OpenAPI-driven development and client code generation (e.g., Orval)
• Experience with a monorepo tooling setup (e.g., pnpm workspaces)
• Familiarity with third-party auth providers (e.g., Clerk, Auth0, Firebase Auth) and multi-tenant access patterns
• Interest in or coursework related to health-adjacent software, privacy-by-design, or HIPAA-aligned system design
• An interest in UX for professional/clinical tools, including designing for a non-technical end user under time pressure