BonaVia Compass

• The description states that Compass is a free bilingual education-planning experience for students and families navigating North American education.
• It uses a structured questionnaire covering student context, interests, academic position, experiences, and goals.
• The backend processes this input into a family-readable public v3/Lin report, which includes four sections:
  • Profile — what the current evidence supports;
  • Direction — a selected priority or clearly labelled working inference;
  • Projects — practical ways to test that direction through real work;
  • Timeline — a clear next step and an actionable sequence.
• The system separates directly provided data, system inferences, and unknowns.
• Reports can be viewed in Chinese or English, and saved as PDF via browser.
• It is described as not predicting admissions outcomes or choosing a future for the student.

Note

The author states that V5 and V5.1 are not part of the submitted product or demo, and only v3/Lin is used in this submission.

• The author positions Compass as an attempt to democratize strategic education planning, contrasting it with high-cost consulting packages.
• It is framed as a tool that helps families make decisions about what is worth the student’s time—and why—not just a list of activities.
• The core claim is that personalized education has become a luxury, and Compass aims to make it basic infrastructure for every child.
• The author emphasizes that Compass is not about giving information, but about connecting dots and turning scattered facts into direction.
• During the hackathon, the product was extended with AI tools (GPT-5.6, Codex) to improve system reliability and reduce race conditions in report generation.

Inference This suggests a shift from idea to prototype, with an emphasis on building a scalable, reliable system using AI-assisted development rather than purely manual design.

• The stated target is students and families navigating North American education.
• It is designed for equally capable students who do not have access to full strategic support, contrasting them with affluent families who can afford multi-year planning systems.
• The tool is described as being used by parents or students themselves, completing a structured questionnaire.
• There is no mention of institutional users (e.g., schools, counselors), though the author notes that the system could be extended into a parallel education layer.

Not evidenced No specific demographic breakdown, age range, or geographic targeting beyond "North American." No evidence of segmentation or customer personas beyond general user types.

• The product is described as free.
• There is no mention of monetization strategies, paid tiers, or subscription models.
• The author mentions a nonprofit foundation but does not describe how this relates to the business model or funding structure.
• No pricing data, revenue streams, or customer acquisition costs are provided.

Inference If the tool remains free, it may rely on non-financial value creation (e.g., community impact, feedback loops) or future monetization through extended services or partnerships.

• The system uses:
  • Browser JavaScript and Node.js serverless handlers.
  • Supabase as persistence layer.
  • OpenAI API and OpenClaw for report generation.
• Report generation can follow either an OpenAI API path or an OpenClaw-orchestrated worker path, both behind quality checks.
• A key technical mechanism is a timestamped generation lease to prevent stale writes from overwriting newer results.
• The judge preview runs on localhost (127.0.0.1), uses fictional data, and makes zero external requests.
• GitHub Actions are used for verification, with 227/227 automated tests passing.
• Codex and GPT-5.6 were used to implement features, isolate changes, and reason through asynchronous failure paths.

Inference The use of AI tools like Codex and GPT-5.6 indicates a rapid prototyping approach, likely driven by the author’s lack of coding background.

• Nine families used early versions of Compass.
  • Feedback indicated that reports were:
    • Personal
    • Organized
    • Professional
    • Practical
  • However, users also noted that reports were too long and needed clearer reading, more specific activities, and better resource recommendations.
• The author states that V5 is being rebuilt based on this feedback.
• No evidence of actual revenue, customer base, or usage metrics beyond the initial nine testers.
• The submitted version uses v3/Lin; V5 is not included in the demo.

Absence of evidence

No data on user retention, engagement, or long-term adoption. No evidence of product-market fit beyond early feedback.

• The author claims Compass is different from most education AI tools, which tend to answer questions or generate recommendations.
• Compass is positioned around the problem of making thoughtful, individualized education strategy more accessible without erasing uncertainty or student agency.
• No direct competitors are named in the description.
• The tool does not appear to be part of a broader marketplace or platform ecosystem.

Inference It may compete with generic AI education tools or high-end consulting firms but lacks clear positioning against specific platforms or services.

• The submitted judge path is local and synthetic, not production-scale.
• The offline preview does not exercise production persistence, delivery, or remote model calls.
• Only fictional data is used in the judge flow.
• V5 and V5.1 are not part of the demo; they are under development.
• The author has no coding background, which raises questions about long-term maintainability or scalability.
• No evidence of external validation, user testing beyond early adopters, or product-market fit.

Red flag

The lack of real-world usage and production-scale testing suggests a high risk of misalignment between the prototype and actual needs.

1. What are the key learnings from the nine families who tested v3/Lin, and how have those shaped V5?
2. Is there any plan to monetize Compass beyond its current free model? If so, what is the strategy?
3. How does the author intend to scale the human review loop as usage grows?
4. What are the risks of relying on AI for inference in a sensitive domain like education planning?
5. Are there any plans to integrate with existing educational institutions or platforms?
6. Has the product been tested in real-world settings beyond the initial feedback loop?