OpenAI 2026 hackathon

Maybe Tomorrow.

An anti-planner for people who need help doing one less thing.

Solo project by Yoshie Yamada · 0 likes · 0 comments

Archive position — measured, not model output

0 likes on Devpost

2,264 of the 7,856 archived projects have more likes, and 5,592 share exactly 0 — so this project's #5,193 place in the like-ranked listing is a tie-break inside that group, not a ranking.

Projects (log scale)

1
10
100
1k
10k
05,592
11,758
2285
3–4132
5–975
10+14

Likes on Devpost. ▲ marks this project's group.

Show the figures
LikesProjectsShare of archive
05,59271.2%
11,75822.4%
22853.6%
3–41321.7%
5–9751.0%
10+140.2%
Devpost like counts for all 7,856 archived projects, captured when this archive was built.

Executive Summary

What the company appears to be: Maybe Tomorrow is a self-reported anti-productivity tool built by one person (Yoshie Yamada) using AI partners — specifically GPT-5.6 Sol and Codex — to design and implement a browser-based application that helps users decide whether to do one activity today or postpone it.

What changed: The project was submitted as part of an OpenAI hackathon, with no evidence of prior development or commercial traction. It is described as a single-person effort using AI tools for product definition, implementation, testing, and documentation.

Single most important open question: Is the tool’s deterministic engine — which evaluates user inputs against a fixed set of rules — sufficient to support meaningful decision-making in real-world contexts, or does it fail to account for nuanced personal circumstances?

Note: This analysis is based solely on the self-reported description provided by the author. No independent verification, revenue data, customer base, or traction evidence exists.

Back to contents

What The Product Actually Is

The description states that Maybe Tomorrow is a tool designed to help users evaluate whether they should do one proposed activity today or postpone it.

  • It allows users to name an activity and answer seven five-point questions plus one about making the idea smaller.
  • A deterministic engine returns one of three verdicts (yes, maybe, no), explaining leading factors.
  • The core interaction takes under a minute and requires no calendar.
  • Optional calendar integration supports importing .ics or Google Calendar ZIP files, parsed entirely in-browser without uploading data.
  • It includes features like a “Today Map” that reports occupied time, overlapping time, protected recovery, and longest continuous opening.
  • For a proposed activity, it shows what would have to change based on the user’s own authorized calendar changes (e.g., Must stay, Can move).
  • Completed decisions can be stored in a browser-local Decision Journal.

Claim: The product is built with AI partners.

  • Inferred from author's description of using GPT-5.6 Sol and Codex.
  • Not evidenced: whether this was a prototype or full product, or if it has been used by others beyond the creator.

Back to contents

Positioning & Claim Evolution

The project positions itself as an anti-planner — not a productivity tool but one that challenges the idea of needing to do everything. It is described as treating postponement as an intentional decision rather than a failure.

  • The tagline: “An anti-planner for people who need help doing one less thing.”
  • The inspiration comes from personal experience with overcommitment and emergency room visits.
  • The product makes no medical claims, but it does frame its purpose in terms of reducing overwhelm through structured questioning.

Claim: It treats postponement as intentional.

  • Evidenced in the description.
  • Inferred: This is a philosophical stance rather than a functional feature.

Back to contents

Target Customer & ICP

The author describes herself as:

  • A novelist, freelance writer, parent, and independent creator.
  • Someone who has faced burnout from trying to carry too much at once.

There is no explicit segmentation of target customers beyond this personal profile. The tool appears designed for individuals seeking clarity about their daily commitments — particularly those who are already overwhelmed or overcommitted.

Claim: The product targets people who struggle with overcommitment.

  • Evidenced in the narrative.
  • Inferred: Likely aimed at creators, freelancers, and knowledge workers who feel pressure to do more.

Back to contents

Business Model & Pricing Evidence

There is no evidence of any business model or pricing structure. The project is described as a hackathon submission with no mention of monetization, subscriptions, or paid features.

Claim: No commercial model is evident.

  • Not evidenced.
  • Inferred: Likely non-commercial due to lack of revenue or pricing data.

Back to contents

Technical & Delivery Signals

The application was built using:

  • AI tools (GPT-5.6 Sol and Codex)
  • React, TypeScript, Vite, Vitest
  • Browser-local processing with no backend or cloud storage
  • Local-first privacy model
  • Deterministic scoring engine
  • Full test coverage (71 tests across 10 files)

Claim: The product uses AI in its development.

  • Evidenced in the description.
  • Inferred: This is a unique approach to software creation involving human-AI collaboration.

Back to contents

Traction & Maturity Signals

There is no evidence of traction, adoption, or user base. The project was submitted as a hackathon entry and lacks any data on usage, retention, or feedback from users beyond the creator.

Claim: No traction or maturity signals are evident.

  • Not evidenced.
  • Inferred: This is an early-stage prototype with no external validation.

Back to contents

Competitive Context

The description does not reference competitors. However, it implies a niche within anti-productivity or decision-support tools — possibly overlapping with time management apps or mindfulness-based planners that encourage slowing down.

Claim: No competitive landscape is described.

  • Not evidenced.
  • Inferred: Likely in a small or emerging segment of the market.

Back to contents

Key Risks & Red Flags

  • Over-reliance on deterministic logic: The system avoids AI advice but may not capture complex personal situations.
  • Limited scope and testing: Built by one person, tested only by the creator; no third-party validation.
  • Privacy model is local but still raises questions: While no data is uploaded, the tool does parse calendar files — a sensitive operation.
  • No commercial viability or scalability: No evidence of monetization, growth plans, or market fit beyond the creator’s experience.

Inference: The product may not scale well without significant refinement.

  • Based on lack of external testing and limited scope.

Back to contents

Diligence Questions To Ask The Founders

  1. How does the deterministic engine handle edge cases that aren’t covered in the current test suite?
  2. What are the limitations of the current calendar parsing, especially around recurrence and exceptions?
  3. Has there been any external user feedback or usability testing beyond the creator’s own use?
  4. Are there plans to expand beyond the current single-activity model?
  5. How would you handle situations where a user wants to do something that conflicts with their own stated preferences?

Back to contents

Investment/Partnership Verdict

Not evidenced: There is no evidence of revenue, customers, or investment interest.

Inference: Early-stage prototype with potential for further development, but not yet ready for commercialization or partnership consideration. The product shows innovation in AI-assisted development and a unique anti-productivity positioning, but lacks traction, scalability, or clear monetization strategy.

This is a speculative read based on the author’s self-description.

Back to contents

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.