OpenAI 2026 hackathon

Requiem

A control room for AI ensembles, built by an AI ensemble. Every action gated by its human owner, every file signed, every event hash-chained. Your machine. Your agents. Your rules.

Solo project by Mitchell Carroll · 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 #6,378 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

Requiem is a desktop application for coordinating user-authenticated AI command-line tools through one interface. The author states it was built by a single human (Mitchell Carroll) with a group of AI collaborators, and that it focuses on Codex CLI in its current version.

What changed

The project began as a personal troubleshooting solution after the founder's Windows installation failed. It evolved into a system for managing AI agents working on CLI tasks, with features like persistent seats, War Room coordination, ApprovalGate execution, and Build Ledger audit trails.

Single most important open question

Does Requiem have any commercial traction or revenue yet? The description states no customers, no revenue, no funding rounds, and no evidence of adoption beyond the author's own development work.

Back to contents

What The Product Actually Is

The description states that Requiem is a desktop application for connecting and coordinating user-authenticated AI command-line tools through one interface. It includes:

  • AI seats with:
    • Callsign
    • Role
    • Model
    • Workspace
    • Access level
    • Color
    • Persistent conversations
  • War Room functionality allowing selected seats to discuss tasks with the user and each other, using a controlled turn structure
  • ApprovalGate that handles execution separately from discussion
  • Build Ledger that records conversations, approvals, activity, and which seat was responsible for each action
  • Hash-chained records and Ed25519 signatures for collaborator attribution and tamper evidence

The author states this is a "Build Week" version focused on Codex CLI, with an adapter system designed to support other compatible CLIs later.

Back to contents

Positioning & Claim Evolution

The description states that Requiem was inspired by the founder's personal experience troubleshooting a Windows installation where he had to manually pass AI responses back and forth. This led to the idea of allowing multiple AIs to work together on a task while maintaining coordination and audit trails.

The author claims Requiem is built "by an AI ensemble" but emphasizes that the human remains the Director responsible for product decisions, approvals, testing, and final direction.

The positioning appears to be:

  • A control room for AI ensembles
  • Built by an AI ensemble (but with human oversight)
  • Focused on CLI tools
  • Emphasizing safety, auditability, and coordination

Back to contents

Target Customer & ICP

The description does not identify specific target customers or personas. It states that the author is "the only human working on the project" and that it was built for "user-authenticated AI command-line tools."

The author mentions that the system supports multiple seats using one CLI installation, and that each seat can have separate conversations without sharing context accidentally.

No evidence of specific customer segments or ICP is provided beyond the general category of users who work with CLI tools and want to coordinate AI agents.

Back to contents

Business Model & Pricing Evidence

The description does not contain any information about pricing, business model, revenue streams, or monetization strategy. The author states that he is "the only human working on the project" and that it was built for personal use during a hackathon.

No evidence of commercial relationships, customer contracts, pricing tiers, or payment mechanisms is provided.

Back to contents

Technical & Delivery Signals

The description indicates Requiem was built with:

  • CSS
  • Python
  • Radix
  • React
  • Rust
  • Tailwind
  • Tauri
  • TypeScript
  • Vite

Key technical features mentioned include:

  • Sequential coordination in the War Room (models receive information and tasks one at a time)
  • Local hash-chained Build Ledger
  • Ed25519 signatures for artifact attribution
  • ApprovalGate for action execution
  • Persistent seats with separate conversations
  • Separation between discussion, trusted instructions, tool output, and approved execution

The author notes that the system uses existing CLI authentication rather than collecting API keys or storing provider credentials.

Back to contents

Traction & Maturity Signals

The description states that:

  • The author is "the only human working on the project"
  • It was built for a hackathon (Build Week)
  • No customers, revenue, or traction data are available
  • The project is planned for release under MPL-2.0 license
  • The author has not yet released any commercial version

There is no evidence of user adoption, customer feedback, product usage metrics, or market validation beyond the author's own development work.

Back to contents

Competitive Context

The description does not provide information about competitors or competitive positioning. It mentions that Requiem supports "other compatible CLIs later" but does not identify existing solutions in this space.

No evidence of direct competitors, market size, or competitive advantages is provided.

Back to contents

Key Risks & Red Flags

  • Single-person development: The author states he is the only human working on the project, which raises questions about scalability and long-term maintenance
  • No commercial traction: No evidence of customers, revenue, or adoption beyond personal use
  • Limited scope: Current focus is on Codex CLI with adapter system planned for later
  • Unverified claims: The "AI ensemble" claim appears to be self-reported without independent verification
  • Hackathon origin: The project was built for a hackathon, suggesting it may not have matured beyond prototype stage

Back to contents

Diligence Questions To Ask The Founders

  1. What is the actual timeline from concept to current state? How long has this been in development?
  2. Are there any early users or beta testers who could validate the need for this product?
  3. What specific problems are users experiencing with existing CLI tool coordination that Requiem solves?
  4. How does Requiem plan to monetize its platform, if at all?
  5. What is the roadmap for expanding beyond Codex CLI support?
  6. How does the system handle cross-agent prompt injection or other security concerns?
  7. What are the actual technical challenges in scaling this approach to larger teams or organizations?

Back to contents

Investment/Partnership Verdict

Confidence: Low

The description provides no evidence of commercial traction, revenue, customers, or funding. The project appears to be a personal hackathon effort with no demonstrated market validation.

The author states he is "the only human working on the project" and that it was built for a hackathon, suggesting it has not yet reached a product-market fit stage.

There are no evidence-based indicators of commercial viability, customer demand, or competitive positioning beyond the author's own claims.

Verdict Not ready for investment or partnership consideration at this time. The project lacks demonstrated traction and commercial evidence to support any meaningful due diligence.

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.