SIGIL — Drawings You Can Run

The description states that SIGIL is a browser-based TypeScript application built with Next.js, React, Three.js, and OpenAI APIs. It interprets uploaded drawings using GPT5.6 Sol to generate structured representations (WorldSpecs), which are then executed by a deterministic simulation engine written in TypeScript.

Key components include:

• A particle rendering system that lifts visual marks from the source image into 3D space.
• A simulation engine handling production, consumption, storage, dependencies, events, metrics, snapshots, replay fingerprints, rewind, and branch comparison.
• Natural language processing for interpreting user interventions via GPT5.6 Sol.
• An architecture supporting multiple world types (Mars Colony, Living Cell, Watershed Ecosystem, Arcane Realm).

The product is described as a "browser-based" tool with no account required to use the public demo.

Evidence Self-reported by author; not independently verified.

The author positions SIGIL as a system for turning drawings into executable simulations — specifically stating that “a drawing was not just a picture of a system, but the first layer of programming it.”

Key claims:

• The ink lifts off the page and becomes a world.
• Drawing is transformed into a living simulation with resources, dependencies, failures, decisions, and consequences.
• Users can trigger crises, intervene, rewind, and compare alternate futures.
• Natural language interventions are converted into bounded actions.
• Rewind and Echo features allow for fair comparison of different paths.

The positioning evolves from a conceptual idea ("What if a drawing was not just a picture...") to a functional prototype with working demos (e.g., Mars Colony).

Evidence Self-reported; no external validation or market positioning data provided.

The description does not clearly define target customers or ideal customer profiles. However, the author implies use cases involving:

• Educational contexts (students sketching systems like colonies, cells, watersheds).
• System design and simulation environments.
• Researchers or educators interested in exploring system behavior through visual means.

There is no mention of specific personas, buyer roles, or institutional adoption.

Evidence Not evidenced; inferred from context of educational and research use cases.

No evidence of a business model or pricing structure is provided. The author mentions:

• Public demo works without an account.
• Potential future features include teacher-created lessons, assignments, shareable capsules, SDKs, etc.
• No indication of monetization strategy, subscription tiers, or paid access.

Evidence Not evidenced; no commercial information available.

SIGIL is built using:

• Frontend: Next.js, React, Three.js, WebGL2, Zustand, Zod
• Backend/Processing: TypeScript, OpenAI Responses API, Codex, GPT5.6 Sol
• Rendering: GPU-driven particle system with custom logic for visual continuity
• Simulation: Deterministic runtime engine managing state, resources, events, replay, and branching

Key technical elements:

• Source ink Awakening uses custom GPU particle rendering.
• Simulation engine handles fixed-step execution, snapshots, replay fingerprints, rewind, and branch comparison.
• GPT5.6 Sol interprets drawings and natural language interventions but does not execute code directly.
• Model outputs are validated before entering the simulation.

The author notes hundreds of passing unit tests, end-to-end browser coverage, public Vercel deployment, and bundled demonstration functionality.

Evidence Self-reported; no independent verification or performance metrics.

There is no evidence of traction or maturity beyond the hackathon submission. The description states:

• It's a demo built during a 4-day hackathon.
• Public demo exists but requires no account.
• No mention of users, customers, revenue, or usage data.
• No indication of product-market fit or user feedback loops.

Evidence Not evidenced; no signs of traction or adoption beyond the author’s own work.

No competitive landscape is described. The author does not reference existing tools or platforms that perform similar functions (e.g., system modeling, simulation software, educational tools). There is no mention of competitors or substitutes in the market.

Evidence Not evidenced; no competitive analysis provided.

• Unproven commercial viability: No evidence of revenue, customers, or monetization.
• AI dependency without control: Reliance on GPT5.6 Sol for interpretation and intervention, which may introduce inconsistency or failure points if not fully validated.
• Limited scalability: The project is described as a single-person effort built in 4 days, suggesting limited scalability or robustness.
• Technical complexity risks: Complex integration of AI with deterministic simulation raises potential issues around accuracy, consistency, and maintainability.
• No clear path to market: No evidence of product-market fit, user feedback, or go-to-market strategy.

Evidence Inferred from lack of commercial data, technical dependencies, and project scope.

1. What specific use cases are you targeting beyond the hackathon demo?
2. Have you tested SIGIL with real users or educators? If so, what feedback did you receive?
3. How do you plan to validate AI interpretations in production environments?
4. Is there any roadmap for monetization or commercial partnerships?
5. What is your strategy for scaling beyond a single-person development effort?
6. Are there any known limitations or edge cases in how the system handles ambiguous or complex drawings?
7. How does SIGIL handle model hallucinations or incorrect interpretations that could lead to invalid simulations?

Evidence Inferred from gaps in self-reported information.