Archive position — measured, not model output
1 like on Devpost
506 of the 7,856 archived projects have more likes, and 1,758 share exactly 1 — so this project's #1,270 place in the like-ranked listing is a tie-break inside that group, not a ranking.
Projects (log scale)
Likes on Devpost. ▲ marks this project's group.
Show the figures
| Likes | Projects | Share of archive |
|---|---|---|
| 0 | 5,592 | 71.2% |
| 1 | 1,758 | 22.4% |
| 2 | 285 | 3.6% |
| 3–4 | 132 | 1.7% |
| 5–9 | 75 | 1.0% |
| 10+ | 14 | 0.2% |
Executive Summary
What the company appears to be
K-Notice is a self-reported tool that processes Korean notices (e.g., apartment, daycare, school, hospital, utility, government) into clear, evidence-backed action plans. It allows users to paste Korean text or take a photo of a notice, and it uses OCR and AI to extract actionable items with citations from the original source.
What changed
The project was submitted as part of the OpenAI 2026 hackathon. The description indicates it is a proof-of-concept product built using a mix of web and mobile technologies, including Astro, Flutter, NestJS, Cloudflare Workers, AWS EC2, and OpenAI Codex.
Single most important open question
Is there any evidence of user adoption or revenue generation beyond the hackathon submission?
What The Product Actually Is
The description states that K-Notice turns Korean notices into clear action plans. It supports multiple notice types (apartment, daycare, school, hospital, utility, government) and allows input via text paste or photo upload.
- OCR processing: On-device OCR is used to avoid sending original photos to servers.
- AI analysis: The system uses OpenAI Codex for processing, with schema-forced output and evidence validation.
- Privacy features: Text masking, local-only handling, deletable history, and local reminders are implemented.
- Output format: Each extracted item (action, fee, warning) must cite an exact substring from the source notice. Unsupported claims are dropped and surfaced as warnings.
The system includes both a web interface and a mobile app (Flutter-based), with a backend service built on AWS EC2 and NestJS.
Inference The product is designed to reduce ambiguity in Korean notices for non-Korean speakers, especially foreign residents, by structuring information into actionable steps.
Positioning & Claim Evolution
The author states that K-Notice began with the idea of turning a notice into a plan you can trust—not just another block of translated text. The core positioning is to provide clarity and trust in complex Korean notices through structured output and evidence-based claims.
Key claims
- A notice should become a plan you can trust.
- Every action, fee, item, and event must cite an exact substring from the source notice.
- Provenance (where each claim comes from) is visible in the interface.
The project evolved from a hackathon submission into a functional prototype with privacy-preserving features and multilingual support.
Inference The positioning emphasizes trust, clarity, and evidence-based AI output over generic translation or summarization tools.
Target Customer & ICP
The description states that K-Notice is aimed at foreign residents in Korea who struggle to understand everyday notices due to language barriers.
Target customer profile
- Foreign residents living in Korea
- Users who encounter Korean notices regularly (e.g., apartment, school, government)
- People seeking clear, actionable information from complex or ambiguous notices
Inference The ICP is narrow and localized—specifically foreign residents navigating Korean bureaucracy or services.
Business Model & Pricing Evidence
The description does not contain any evidence of a business model or pricing structure. It only describes the product's functionality and technical implementation.
Not evidenced
Technical & Delivery Signals
The system uses:
- Web stack: Astro, Cloudflare Pages, Cloudflare Worker, D1, Workers AI, Tesseract.js
- Mobile app: Flutter, backed by NestJS service on AWS EC2
- AI engine: OpenAI Codex (schema-forced output and evidence validation)
- Data handling: On-device OCR, local masking, server-side text-only processing, PostgreSQL job queue
- Security & privacy: Ephemeral sandboxed execution, deletable history, local-only reminders
Inference The architecture suggests a focus on privacy, scalability, and structured AI output. The use of schema-forced output and evidence validation implies an emphasis on trustworthiness over generality.
Traction & Maturity Signals
The description does not include any evidence of traction or adoption beyond the hackathon submission.
Not evidenced
Competitive Context
No competitive analysis is provided in the description. The author does not mention existing tools or platforms that address similar needs (e.g., notice translation, action plan generation).
Not evidenced
Key Risks & Red Flags
- No revenue or traction evidence: The product exists only as a hackathon submission with no indication of monetization or user base.
- Limited scope: The focus is on one specific use case (Korean notices) and one language (Korean), which may limit scalability.
- Dependency on AI model: Reliance on OpenAI Codex for processing introduces risk if the API changes or becomes unavailable.
- Privacy vs. usability trade-offs: On-device OCR and local-only handling are good privacy features, but may limit performance or accuracy.
Inference The project is in a very early stage with no commercial evidence. It is unclear whether it has moved beyond prototype or if there is any market demand.
Diligence Questions To Ask The Founders
- Has the product been tested with real users outside of the hackathon?
- Are there plans to monetize the service, and what is the proposed business model?
- How does the system handle edge cases in notice formats or content that may not fit the schema?
- What are the long-term plans for scaling beyond the current tech stack (e.g., OpenAI API migration)?
- Is there any internal data or feedback from users that shows adoption or usage patterns?
Investment/Partnership Verdict
The description indicates K-Notice is a hackathon project with no evidence of traction, revenue, or customer base. It is not clear whether the product has moved beyond prototype or if it has been validated in real-world use.
Not evidenced
Source
Submitted to the OpenAI 2026 hackathon on Devpost. Project home on DevPost.
The analysis above was generated by a language model from the project's own one-line description. It is not independent research and contains no verified traction, revenue or customer data.