OpenAI 2026 hackathon

Fleetic Keyboard

An Android keyboard that refines workplace messages with GPT-5.6 while preserving the way you naturally communicate.

Solo project by Yanuar Arifin · 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 #4,138 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

Fleetic Keyboard is an Android keyboard application that integrates GPT-5.6 to refine workplace messages while preserving the user's communication identity. It is built as a native Android input method, leveraging AOSP LatinIME and a privacy-conscious backend. The product allows users to define a "Communication DNA" profile based on writing samples and refinement priorities, which guides AI-assisted text refinement.

What changed

The author describes Fleetic as a personal project developed during an OpenAI hackathon (Build Week). It is presented as a proof-of-concept with a focus on trust, privacy, and user control. The product does not appear to have launched commercially or gained traction beyond its developer's own use.

Single most important open question

Is there any evidence of commercial viability, user adoption, or revenue generation? The description contains no data on users, customers, monetization, or market traction — only a self-reported technical and conceptual framework.

Back to contents

What The Product Actually Is

The description states that Fleetic is a native Android keyboard built using AOSP LatinIME as its foundation. It integrates GPT-5.6 to refine workplace messages in real time, but only after the user explicitly taps "Refine". Key features include:

  • Communication DNA: A structured profile derived from a writing sample and refinement priority.
  • Refinement is intentional: text is not sent until the user confirms.
  • Text ownership is conservative: the keyboard replaces only what it can prove it owns.
  • The system fails safely if replacement cannot be verified.

The product is described as an input method that works within existing Android apps (e.g., Slack, email clients), and does not replace the entire messaging experience.

Claim

Fleetic is a native Android keyboard with AI-assisted refinement.

Evidence Author's own description.

Back to contents

Positioning & Claim Evolution

The author positions Fleetic as an alternative to generic grammar checkers or chatbot keyboards, emphasizing:

  • User voice preservation: The output should sound like the user.
  • Intentional sending: No automatic replacement; users must tap Refine.
  • Privacy-first design: No message content is stored or logged.
  • Trust boundary: Text replacement only happens when ownership can be verified.

The positioning evolves from a technical solution to a trust-based communication tool, where the AI’s role is to assist without altering identity.

Claim

Fleetic aims to reduce hesitation before sending messages by refining them naturally, while preserving user voice.

Evidence Author's own description.

Back to contents

Target Customer & ICP

The author describes the target audience as:

  • Professionals working in English who are not native speakers.
  • Users who experience hesitation or context-switching when typing short workplace messages (e.g., Slack, email).
  • Individuals who want AI help but do not want to lose their communication identity.

There is no explicit segmentation beyond this general user group. No named customer types, personas, or use cases are provided.

Claim

Fleetic targets professionals working in English as a second language.

Evidence Author's own description.

Back to contents

Business Model & Pricing Evidence

The description does not contain any information about:

  • Revenue streams
  • Pricing models
  • Monetization strategy
  • Customer acquisition costs
  • Sales process or go-to-market plan

It is unclear whether the product is intended to be free, paid, or offered via subscription.

Claim

No business model or pricing evidence provided.

Evidence Not evidenced.

Back to contents

Technical & Delivery Signals

The author provides detailed technical information:

  • Built on AOSP LatinIME
  • Uses Kotlin, Java, C++, Jetpack Compose, and Android Views
  • Backend in Go with Chi, PostgreSQL (via pgx), Distroless containers
  • Communication DNA is structured into six dimensions: brevity, formality, warmth, directness, vocabulary, expressiveness
  • GPT-5.6 is used via OpenAI’s Responses API
  • Uses strict schema validation and idempotent request handling
  • Includes tests covering contracts, handlers, prompt invariants, and Android API client behavior

The product was tested on a physical Samsung Galaxy A54 running Android 16.

Claim

Fleetic is built with strong technical rigor, including privacy-conscious design and safety mechanisms.

Evidence Author's own description.

Back to contents

Traction & Maturity Signals

There is no evidence of:

  • User adoption or retention
  • Revenue or monetization
  • Customer feedback or usage metrics
  • Product roadmap or iteration history beyond the MVP
  • Any form of market validation

The project is described as a Build Week hackathon submission, and no mention of launch, beta testing, or user engagement.

Claim

No traction or maturity signals.

Evidence Not evidenced.

Back to contents

Competitive Context

The author does not reference any competitors directly. However, the product’s positioning implies it competes with:

  • Generic AI keyboards
  • Grammar-checking tools (e.g., Grammarly)
  • Chatbots integrated into messaging apps

It distinguishes itself by focusing on user identity preservation, intentional refinement, and trust boundaries.

Claim

Competes in the AI-assisted writing space, with a focus on trust and identity.

Evidence Author's own description.

Back to contents

Key Risks & Red Flags

Several risks are implied by the description:

  • No commercial traction or monetization: The product is not yet launched or monetized.
  • Limited scope: Only supports Android; no iOS or desktop plans mentioned.
  • High technical complexity: Integrating AI into a keyboard with trust and safety constraints may be difficult to scale.
  • Unproven user need: No evidence of market demand or user feedback beyond the developer’s own experience.
  • Dependency on GPT-5.6: The system relies heavily on a proprietary model, which could introduce risks if access changes.

Claim

Risks include lack of traction, limited scope, and unvalidated assumptions.

Inference Based on absence of evidence for commercial viability or user feedback.

Back to contents

Diligence Questions To Ask The Founders

  1. What is the expected timeline to market?
  2. How will you validate that users actually want this product?
  3. Are there any plans to expand beyond Android?
  4. What are your assumptions about user behavior and adoption?
  5. How do you plan to handle model drift or degradation over time?
  6. Is there a plan for user onboarding or education around Communication DNA?
  7. What are the key metrics you would track post-launch?

Claim

These questions are necessary due to lack of evidence around traction, user validation, and commercial strategy.

Inference Based on absence of data.

Back to contents

Investment/Partnership Verdict

There is no evidence of:

  • Revenue or ARR
  • Customer base or adoption metrics
  • Product-market fit or user feedback
  • Commercial readiness or scalability
  • Funding or investor interest

The project is described as a self-developed MVP for a hackathon, with no indication of commercial intent or traction.

Claim

No evidence of investment or partnership viability.

Evidence Not evidenced.

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.