Onoki

This analysis is based entirely on the self-reported, unverified description provided by the author. No third-party verification or historical data is available.

• The description states that Onoki is an app with support for on-device user profiles.
• Users can create learning topics and goals per profile.
• Sessions are initiated either for a topic/goal or in free mode, starting an agent loop with speech input/output optimized for low latency.
• The system works transport agnostic across SST->LLM->TTS engines and supports real-time sessions via OpenAI Realtime API and Gemini Live API.
• The agent has tools for tracking learning progress, memory about the user, suggesting goals, and creating/updating visual scenes using a custom, token-efficient DSL.
• The DSL supports state, animations, geometry constructions, and charts/diagrams.
• Onoki uses the Socratic method to guide learners toward understanding.

Not evidenced: No information on actual product functionality beyond the author's description. No evidence of real-world usage or performance metrics.

• The author claims Onoki is an AI learning companion using speech and visualizations.
• It positions itself as a tool for children and general learners, aiming to make AI more effective in educational contexts.
• The app is described as being built with the Socratic method in mind — guiding users through questions to build true understanding.
• The author mentions that it was inspired by their children using ChatGPT but not automatically functioning as a learning companion.

Not evidenced: No claims about market positioning, differentiation from competitors, or prior versions. No evidence of marketing or branding efforts beyond the project submission.

• The description states that Onoki is intended for children and general learners.
• It supports on-device user profiles, implying personalization for individual users.
• Sessions can be tailored to specific learning topics or goals, suggesting a focus on structured learning experiences.

Not evidenced: No evidence of actual target customers, customer segments, or personas. No data on who uses the app or how many people are using it.

• The description does not mention any pricing model or monetization strategy.
• There is no indication of whether Onoki will be sold as a paid product, offered for free, or supported through other means (e.g., subscriptions, ads).

Not evidenced: No evidence of business model, pricing plans, or revenue streams.

• Built with Rust, Swift, Tauri, and TypeScript.
• Uses a custom DSL tailored to learning situations.
• Supports real-time sessions via OpenAI Realtime API and Gemini Live API.
• The system works transport agnostic across SST->LLM->TTS engines.
• A test harness was developed for end-to-end validation of the DSL.
• The author used GPT-5.6 to optimize the DSL during development.

Not evidenced: No evidence of technical scalability, deployment infrastructure, or delivery mechanisms beyond the developer's own setup.

• The project is described as being in TestFlight, suggesting early-stage release.
• It was submitted to the OpenAI 2026 hackathon.
• The author has been working on it since Opus 4.5 came out.
• A test harness and end-to-end validation process were built, which the author considers a significant improvement.

Not evidenced: No evidence of user adoption, retention, or usage statistics. No data on product maturity or long-term development plans.

• The description does not compare Onoki to existing learning tools or AI companions.
• It is implied that current solutions like ChatGPT are insufficient for educational purposes without additional features.
• No mention of competitors or market analysis.

Not evidenced: No evidence of competitive landscape, market positioning, or awareness of similar products.

• The project is built by a single person (Stephan Aßmus), raising concerns about scalability and long-term maintenance.
• The use of a custom DSL may limit flexibility and increase complexity in implementation.
• The system trades low latency for reduced ability of the model to map tasks onto outputs, which could affect performance.
• No evidence of funding, team expansion, or commercial traction.

Not evidenced: No evidence of financial backing, team growth, or strategic partnerships.

1. What specific learning outcomes have you observed from using Onoki?
2. How do you plan to scale beyond a single developer?
3. Are there any users currently testing the app, and what feedback have they provided?
4. What is your roadmap for monetization or commercial viability?
5. How do you intend to ensure quality and consistency of learning experiences across different LLMs?

These questions are based on the lack of evidence around traction, scalability, and business strategy.