Paper Orbit

What the company appears to be: Paper Orbit is a self-reported daily paper-reading companion for researchers that personalizes arXiv paper recommendations using a hybrid local/public recommender system, integrates with OpenAI-compatible APIs for PDF reading and analysis, and claims to keep all user data on-device. It was built as a submission to the OpenAI 2026 hackathon.

What changed: The project is described as a personal solution to the problem of information overload in research, where users are overwhelmed by hundreds of new arXiv papers each day. It introduces a system that avoids traditional recommendation trade-offs (e.g., keyword alerts or centralized recommender services) by using a public candidate pool and local ranking.

Single most important open question: Is there any evidence of user adoption, revenue, or traction beyond the hackathon submission?

The description states that Paper Orbit is a daily paper-reading site for researchers. It offers:

• Ten personalized paper recommendations every day.
• Structured arXiv search.
• A "Paper Copilot" that answers questions grounded in the full PDF.
• Auto-generated reading reports.
• A personal library with reading progress.

It requires users to sign in via ChatGPT and connect their own OpenAI-compatible API key. The system is built using Next.js + TypeScript, deployed on Sites, and uses a two-layer recommender system: public candidate generation and local ranking.

Evidence: The author states this is the product’s function.

Inference: It appears to be an early-stage tool for academic researchers, not yet commercialized or monetized. No revenue or customer data is provided.

The description positions Paper Orbit as a personal, privacy-preserving alternative to existing arXiv tools. The key claims are:

• It avoids keyword alerts that don’t know the user.
• It avoids recommender services that know too much (i.e., centralized systems).
• It reads full PDFs, not just abstracts.
• It keeps the research profile entirely on the user's device.

The project evolved from a question: “Can a daily paper companion be genuinely personal, read the actual PDF, and still keep your research profile entirely on your own device?”

Evidence: The author states this is the product’s positioning and evolution.

Inference: This suggests a strong focus on privacy and user control, but no evidence of how this differentiates it in the market or whether users value these features.

The description states that Paper Orbit is for “anyone who signs in with ChatGPT” and targets researchers. It is built for people who are overwhelmed by arXiv’s daily listing of hundreds of new papers.

Evidence: The author describes the target audience as researchers and users of arXiv.

Inference: The ICP (Ideal Customer Profile) appears to be academic or research professionals, but no segmentation or user data is provided.

The description does not state a business model or pricing structure. It only mentions that users connect their own OpenAI-compatible API key and that the system uses a two-layer recommender with public candidates and local ranking.

Evidence: No explicit mention of monetization, subscriptions, or pricing.

Inference: The project is self-reported as a hackathon submission, suggesting no commercial model exists yet.

The system is built using:

• Next.js + TypeScript
• Deployed on Sites
• Identity via “Sign in with ChatGPT”

Key technical features include:

• A two-layer recommender: public candidate generation and local ranking.
• PDF-grounded copilot that verifies files independently before model calls.
• API key handling via AES-GCM-encrypted HttpOnly cookies.
• Data stored only in localStorage, namespaced per signed-in account.
• Failure diagnostics with correlation IDs but no echoing of secrets.

Evidence: The author describes the architecture and technical implementation.

Inference: The system is designed for privacy and robustness, but there’s no evidence of production deployment or scale.

There is no evidence of user traction, revenue, or adoption beyond the hackathon submission. The project is described as a two-person team effort with no mention of users, customers, or growth metrics.

Evidence: No data on users, customers, or business performance.

Inference: This is an early-stage prototype, not yet a product in use by real users.

The description does not name specific competitors. However, it positions Paper Orbit as an alternative to:

• Keyword alerts (e.g., arXiv’s email alerts)
• Centralized recommender services (e.g., those that store user data)

It also contrasts itself with LLMs that only read abstracts, not full papers.

Evidence: The author describes the problem space and how Paper Orbit addresses it.

Inference: No evidence of competitive analysis or market positioning beyond self-reporting.

• No commercial traction or revenue: The project is described as a hackathon submission with no evidence of monetization.
• Unproven user demand: No data on how many users would actually use the product or value its privacy features.
• Limited scalability assumptions: The system uses local ranking and device-local storage, which may not scale to a broader audience.
• No third-party verification: All claims are self-reported with no independent validation.

Evidence: The project is described as a hackathon submission; no evidence of real-world usage or commercial viability.

1. What is the actual user feedback from early adopters, if any?
2. How do you plan to monetize this product beyond the current prototype?
3. Have you validated the privacy claims with users or security experts?
4. What are the technical limitations of local ranking at scale?
5. Are there any plans for a paid version or premium features?