I’m building a conversational assistant that automatically fills out patient intake forms for our healthcare platform. The bot’s single focus is processing patient forms, not booking appointments or answering general medical questions. Each intake form must capture two data groups every time: • Personal information (name, DOB, contact details, address, etc.) • Medical history (current conditions, medications, allergies, prior surgeries) If you can also future-proof the flow so insurance details could be added later, that’s a plus, but today’s milestone is strictly the two sections above. What I need from you – Design and build an AI-driven dialogue that collects the required fields naturally, validates answers in real time, and summarizes responses in a structured JSON or database-ready format. – Choose the framework you’re fastest with (OpenAI GPT, Rasa, Dialogflow, LangChain, or a comparable stack) and explain why it fits. – Handle quick turnarounds on tweaks; medical compliance and usability reviews often come with same-day feedback. – Supply clean, well-commented code plus brief deployment notes so my dev team can drop the chatbot into our React front end. Acceptance criteria 1. User can complete the full intake conversation in under five minutes with no dead ends. 2. Required fields cannot be skipped; optional notes can be left blank. 3. Output schema matches the JSON sample I’ll share on kickoff. 4. Model responses stay under 2 seconds on average. 5. You provide a short Loom walkthrough and handoff documentation. I have ongoing work for upgrades and new form types once this pilot proves solid, so speed and reliability here will lead to a long-term collaboration.