GeneCourt

GeneCourt is an educational tool built for genetic variant classification training. The description states it presents a courtroom-style interface to teach learners how scientific disagreement arises in clinical genetics, using real ClinVar records and interactive 3D structures. It is described as a strict TypeScript/React application deployed on Cloudflare Workers, with a deterministic pipeline that prevents live model outputs from overriding verified evidence.

The author claims the tool makes uncertainty visible by allowing users to inspect why credible science can disagree, and to see how different types of evidence support or challenge variant classifications. It includes features like interactive 3D protein structures, source tracing, and export capabilities.

Key commercial due-diligence questions include: Is there a clear path from this educational demo to a product that could serve real users? What is the intended market beyond hackathon education? How does it differ from existing scientific data platforms or educational tools?

The most important open question is whether GeneCourt has any evidence of traction, revenue, or adoption beyond its own demonstration. The description contains no information about customers, pricing, usage metrics, or commercialization plans.

The description states that GeneCourt is a "classroom-style evidence lab" where learners can examine genetic variant classifications using real ClinVar data and interactive 3D structures. It presents two curated variants from the TP53 gene domain:

• R282Q, a conflict case with conflicting ClinVar assertions (Likely pathogenic and Uncertain significance)
• R248Q, a calibration control with an expert-panel Pathogenic record

The product allows users to:

• Inspect transcript, coding change, protein change, condition, submitter, and assertion date
• Follow a 90-second replay through four evidence perspectives (supporting, challenging, methodology, population context)
• Focus on affected residue in interactive 3D protein structure
• Open every evidence card's source, identifier, admitted source span, retrieval date, and limitation
• Review what is established, disputed, contextual, missing, and unresolved
• Admit PubMed records and inspect changes
• Export or print case files

The description states it is built with Next.js, React, TypeScript, Zod, Mol*, Cloudflare Workers, and OpenAI GPT-5.6.

The author states that GeneCourt turns genetic variant classification "from a final label into reasoning visible." It claims to make "uncertainty something learners can inspect—not something an AI quietly hides."

The positioning evolved from a hackathon project focused on educational demonstration to one that presents itself as teaching scientific reasoning in clinical genetics. The description emphasizes that it does not diagnose disease, estimate risk, recommend treatment, or autonomously classify variants.

The claim is that it makes "credible science can disagree" visible through interactive evidence inspection and cross-examination. It positions itself as a tool for understanding how scientific disagreement arises rather than resolving it.

The description states GeneCourt is designed for "learners" in genetic variant classification training. It is described as an educational and research demonstration, not intended for clinical use or patient care.

The target audience appears to be:

• Educational institutions teaching clinical genetics
• Researchers studying variant classification methods
• Trainees in genetic counseling or clinical laboratory science

The description does not identify specific customer segments beyond "learners" or specify whether it targets academic, professional development, or institutional users. No evidence of customer personas, usage patterns, or market segmentation is provided.

The description states GeneCourt is an educational and research demonstration that does not diagnose disease, estimate personal risk, recommend treatment, or autonomously classify patient variants. It explicitly states that expert review is always required.

There is no evidence of pricing structure, revenue model, or monetization strategy in the description. The project appears to be a demo-only tool with no commercialization plans described.

The description states GeneCourt is:

• A strict TypeScript and React application using Next.js App Router
• Compiled with Vinext and deployed as Cloudflare Worker
• Uses Zod contracts for data validation
• Renders 3D structures with Mol* locally served PDBx/mmCIF files
• Implements a guarded server-side pipeline using OpenAI Responses API and GPT-5.6
• Has deterministic code checks for source spans, allowlisted hosts, exact variant/condition scope, experimental model, endpoint, safety language, and schema validity
• Uses Codex for research planning, architecture, typed contracts, source adapters, validator design, UI implementation, testing, and submission preparation

The description states it includes keyboard controls, reduced-motion behavior, accessible source drawers, JSON export, print output, and responsive layouts. It passed 108 automated tests plus typecheck, lint, production build, evidence-integrity evaluation, accessibility, desktop/mobile, reduced-motion, export, print, and browser error checks.

The description states GeneCourt is an educational and research demonstration submitted to the OpenAI 2026 hackathon. It does not contain any evidence of traction, revenue, customers, or adoption beyond its own demonstration.

There is no evidence of:

• User base or customer acquisition
• Revenue or monetization
• Product usage metrics
• Market validation or customer feedback
• Commercial deployment or production use

The description indicates it was built as a hackathon submission and remains in demo form.

The description does not provide information about competitive landscape, existing alternatives, or market positioning relative to other tools. It makes no claims about how GeneCourt differs from existing scientific data platforms, educational tools, or genetic variant classification systems.

No evidence of competitors, market share, or differentiation strategy is provided in the description.

• The project is described as an educational demo submitted to a hackathon with no commercialization plans
• No evidence of traction, revenue, customers, or adoption beyond its own demonstration
• No pricing model or monetization strategy described
• No indication of how it would transition from educational tool to commercial product
• The description states it does not diagnose disease or classify variants, limiting its clinical utility
• The deterministic pipeline approach may limit scalability or real-time updates
• No evidence of team experience in commercializing educational or scientific tools

1. What is the intended transition path from this educational demo to a product that could serve real users?
2. How does GeneCourt plan to differentiate itself from existing scientific data platforms and educational tools?
3. What specific market segments are you targeting beyond academic institutions?
4. Are there any plans for monetization or revenue generation?
5. What is the roadmap for expanding beyond the TP53 gene domain?
6. How do you plan to handle educator approval and persistence in live ingestion workflows?
7. What evidence supports the need for this type of educational tool in clinical genetics training?
8. How will you validate that learners improve at distinguishing different types of genetic evidence?