Building an AI Sleep Coach: How Rest is Making CBTI Principles Accessible to DIY Sleep Hackers

30 snips

Nov 20, 2025

Martin Siniawski, CEO of Rest, and Ignacio, CTO, dive into the creation of an AI sleep coach. They discuss how podcast listener behavior illuminated the sleep challenge. The duo explains their choice of voice interaction for warmth and trust in their product, built on clinically proven CBT-I principles. They reveal innovative features like memory systems for personalized user experiences and dynamic agendas for daily check-ins. Plus, they share insights on ethical boundaries in health tech and the importance of expert feedback for iterative improvements.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

ADVICE

Base The Product On CBTI Principles

Use CBTI as a foundation because it has high, clinically proven efficacy and addresses root causes of insomnia.
Digitize CBTI into a guided eight-week protocol to scale access affordably.

INSIGHT

Voice Builds Trust Faster Than Text

Trust and daily support are critical for behavior change, and a cold app struggles to create them.
Voice increases warmth, intimacy, and user openness compared with text-only interactions.

ANECDOTE

Daily Coach-Led Check-Ins

After sleep logging, users do a daily check-in conversation where the AI drives the agenda and can be interrupted.
This 'coach-led' flow mirrors expert practice but at daily frequency impossible for human therapists.

Get the Snipd Podcast app to discover more snips from this episode

Get the app

Guests

Martin Siniawski, CEO and co-founder, Rest
Ignacio, CTO, Rest

You'll hear how they:

Discovered the sleep use case from podcast app user behavior (10% of users, but high willingness to pay)
Used jobs-to-be-done research to identify "DIY sleep hackers" as an underserved segment
Chose CBTI (Cognitive Behavioral Therapy for Insomnia) as their foundation—a clinically proven approach with 80% efficacy
Evolved from text chatbot to voice-first AI using Vapi for voice and OpenAI for reasoning
Built a memory system that remembers user context (like traveling, having a dog) with time-based relevance
Created dynamic agendas that drive daily conversations based on sleep data, program stage, and user compliance
Managed parallel development paths (text via OpenAI Assistants and voice via Vapi)
Moved from massive system prompts to RAG for general sleep knowledge, keeping user data in prompts
Navigated wellness vs. medical product positioning with clear guardrails against diagnosis and medication advice
Used weekly error analysis with domain experts (sleep therapists) to drive product iterations
Built LLM-powered evals for safety boundaries and experimented with Hamming for voice testing

Resources & Links

Rest – AI sleep coach app
Vapi – Voice agent platform Rest uses
Langfuse – Observability and evals platform
Hamming – Voice testing platform
AI Evals Maven Course by Hamel Husain and Shreya Shankar (Get 35% off with Teresa's affiliate link)

Chapters

00:00 Introduction to Rest and Its Founders 00:33 The Origin Story of the AI Sleep Coach 02:07 Exploring the Podcast App and Sleep Use Case 03:35 Transitioning to a Dedicated Sleep Audio App 05:47 Understanding User Segments and Sleep Challenges 07:45 Introduction to the AI Sleep Coach 13:14 The Role of Voice in the AI Sleep Coach 18:46 Daily User Interaction and Features 21:30 Prototyping and Early Learnings 28:09 Navigating Ethical and Regulatory Concerns 30:39 Navigating the Line Between Health and Wellness Apps 31:00 Incorporating Adjacent Disciplines into the App 32:15 The Power of 24/7 Availability 32:53 Evolution of the Chatbot and Error Analysis 34:49 User Experience Improvements and Voice Integration 46:49 Implementing Memory and Personalization 50:18 Dynamic Agenda and User-Centric Conversations 57:37 Evaluation and Guardrails 01:00:05 Future Roadmap and Enhancements 01:03:38 Combining Data Layers for Enhanced AI 01:06:00 Conclusion and Final Thoughts