Archive position — measured, not model output
1 like on Devpost
506 of the 7,856 archived projects have more likes, and 1,758 share exactly 1 — so this project's #856 place in the like-ranked listing is a tie-break inside that group, not a ranking.
Projects (log scale)
Likes on Devpost. ▲ marks this project's group.
Show the figures
| Likes | Projects | Share of archive |
|---|---|---|
| 0 | 5,592 | 71.2% |
| 1 | 1,758 | 22.4% |
| 2 | 285 | 3.6% |
| 3–4 | 132 | 1.7% |
| 5–9 | 75 | 1.0% |
| 10+ | 14 | 0.2% |
Executive Summary
What the company appears to be
COGNISECT is a tool for mathematics teachers to test competing formal explanations for one signed-integer error. It uses GPT-5.6 to map observed student work into a closed rule registry, then applies deterministic code to scan 625 problems and compile one separating probe. The teacher controls both release and interpretation of the probe.
What changed
The project began as a boundary-based design: formal rules can explain answers but not cognitive states. It evolved into an evidence workbench where AI maps observations into a fixed registry, while deterministic code exposes disagreement between procedures. No learner-facing content is released automatically; all decisions are gatekept by the teacher.
Single most important open question
Is there a real-world need for this specific workflow in formative assessment? The description states no revenue, customers or traction data beyond the author's own account.
What The Product Actually Is
The description states that COGNISECT helps mathematics teachers test competing formal explanations for one signed-integer error. It uses GPT-5.6 to map observed student work into a closed rule registry and deterministic code to scan 625 problems and compile one separating probe. The teacher controls both release and interpretation.
Evidence
- "COGNISECT helps a mathematics teacher test competing formal explanations for one signed-integer error."
- "GPT-5.6 maps observed work into a closed rule registry; deterministic code scans 625 problems and compiles one separating probe."
- "The teacher controls both release and interpretation."
Inference COGNISECT is not an AI chatbot or general-purpose tool but a specialized formative assessment tool for educators.
Positioning & Claim Evolution
The description states that COGNISECT began with a boundary, not a chatbot prompt: formal rules can explain answers but cannot prove cognitive state. This led to a different role for AI in formative assessment — one where the model performs constrained mapping and deterministic code exposes disagreement, keeping the teacher responsible for interpretive decisions.
Evidence
- "The project began with a boundary, not a chatbot prompt: a formal rule can explain an answer, but it cannot prove a learner's cognitive state."
- "That suggested a different role for AI in formative assessment. Let the model perform constrained mapping, let deterministic code expose the disagreement, and keep the teacher responsible for every learner-facing and interpretive decision."
Inference The positioning evolved from a general-purpose AI tool to a specific educational workflow that emphasizes transparency, teacher control, and evidence-based reasoning.
Target Customer & ICP
The description states that COGNISECT is designed for mathematics teachers who want to test competing formal explanations for signed-integer errors. The product is built around an educator-authored exemplar or de-identified student work.
Evidence
- "COGNISECT helps a mathematics teacher test competing formal explanations for one signed-integer error."
- "The teacher starts with a provenance-cleared, educator-authored project exemplar or enters de-identified work for one signed-integer subtraction problem."
Inference The primary customer is a mathematics teacher using the tool in a classroom setting. The ICP likely includes educators who value evidence-based assessment and want to avoid premature conclusions about student understanding.
Business Model & Pricing Evidence
Not evidenced.
Explanation
There is no mention of pricing, licensing, or monetization strategy in the description.
Technical & Delivery Signals
The description states that COGNISECT separates generative mapping from executable authority. GPT-5.6 performs structured mapping and drafts a teacher-facing proposal, while deterministic code handles execution. The system uses FastAPI, Python 3.12, PostgreSQL 17, LangGraph, Next.js, React, TypeScript, Vercel, Render, Expo.io, and codex.
Evidence
- "COGNISECT separates generative mapping from executable authority."
- "GPT-5.6 Terra performs the default structured mapping and drafts a teacher-facing proposal."
- "The deterministic core is Python 3.12 with a total rule interpreter and an exhaustive Counterexample Compiler."
- "FastAPI and Pydantic expose strict contracts."
- "SQLAlchemy, Alembic, PostgreSQL 17, and LangGraph checkpoints preserve ownership, idempotency, interrupts, resumes, audit transitions, and deletion."
- "Frontend is Next.js 16, React 19, and strict TypeScript."
Inference The system is built with a clear separation of concerns between generative AI and deterministic execution. It includes robust auditing, concurrency handling, and privacy controls.
Traction & Maturity Signals
Not evidenced.
Explanation
There is no mention of revenue, customers, or adoption in the description. The project was submitted to a hackathon and has no evidence of traction beyond the author's own account.
Competitive Context
Not evidenced.
Explanation
The description does not provide any information about competitors or market positioning beyond the self-reported claims.
Key Risks & Red Flags
- No revenue, customers or traction data: The project is described as a hackathon submission with no evidence of real-world usage.
- Single-person team: The team size is listed as one, which may limit scalability and development capacity.
- Highly specialized use case: The tool is designed for a narrow educational context (signed-integer errors), limiting its potential market reach.
- Dependency on GPT-5.6: The system relies heavily on a specific AI model, which could pose risks if the model changes or becomes unavailable.
Evidence
- "Team size: 1"
- "The project was created during the Build Week submission period; dated commits document the work."
- "Everything above is the authors' own account. It is not independently verified, and no revenue, customer or traction data is available beyond what they state."
Diligence Questions To Ask The Founders
- What is the actual demand for this type of tool in real classrooms?
- How does the system handle edge cases or unexpected student responses?
- Are there plans to expand beyond signed-integer subtraction problems?
- What are the long-term sustainability and scalability plans for the product?
- How do you plan to monetize or commercialize this tool?
Investment/Partnership Verdict
Not evidenced.
Explanation
There is no information provided about funding rounds, valuations, or investment interest in the project. The description does not indicate any commercial traction or investor interest beyond the author's own account.
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.