Between Sessions

Between Sessions is a secure, passcode-gated AI companion web application designed for mental health clients to use between scheduled therapy appointments. The application provides a structured, supportive conversational interface powered by a large language model, allowing users to process thoughts, track their emotional state, and engage in reflective journaling without direct clinician involvement.

Mental health care has a fundamental continuity problem: therapy sessions typically occur once per week, leaving clients to navigate the intervening days without structured support. Research consistently shows that the period between sessions is when coping skills are tested most severely, and when clients who would benefit most from additional support have the least access to it. While crisis hotlines and texting services exist, there is a gap between crisis-level intervention and the low-intensity, reflective support many clients need. AI-powered conversational tools offer a promising approach to filling that gap — providing a non-judgmental, always-available space for clients to process their experiences in a clinically-framed way. Between Sessions is motivated by the belief that this kind of support, when designed thoughtfully and with appropriate guardrails, can meaningfully extend the therapeutic relationship beyond the 50-minute session. The passcode-gating model — where the clinician issues the access code to the client — ensures the tool remains part of a supervised care relationship rather than a standalone consumer product, preserving the clinical context that makes such a companion genuinely useful rather than potentially harmful.

The primary objective of this capstone engagement is to extend the core Between Sessions application with features that increase clinical safety, session continuity, and user engagement. The deliverable will be a functional pilot-ready web application that demonstrates these extensions through working code, tests, and documentation.

• Coursework or demonstrable project experience in modern JavaScript or TypeScript;
• Familiarity with React component architecture and state management;
• Understanding of REST API design and HTTP request/response cycles;
• Exposure to relational databases (SQL) at the level of writing basic queries and understanding foreign key relationships;
• Comfortable working in a Unix-like terminal environment and using Git for version control in a collaborative setting

• Prior experience with React Query or any data-fetching library;
• Familiarity with OpenAI or similar LLM APIs, particularly streaming completions;
• Exposure to an ORM such as Drizzle or Prisma;
• Experience writing Playwright or Cypress end-to-end tests;
• Prior coursework in human-computer interaction, UX design, or software engineering with user-facing systems;
• Given the clinical context of this application, an interest in or background in health informatics, psychology, or accessibility is a meaningful plus;
• The team should include at least one member comfortable with backend API development and at least one with frontend UI/UX sensibility