Visible Thinking Designer

What the company appears to be: Visible Thinking Designer is an AI-assisted design tool for tertiary and vocational educators in Aotearoa New Zealand. The author states it helps tutors redesign real learning tasks so that learner thinking—rather than polished outputs—is made visible and usable as evidence of capability.

What changed: The project emerged from fieldwork with educators, integrating educational research into a self-contained web application built using Next.js, TypeScript, OpenAI APIs, and GPT-5.6. It is described as a public prototype deployed via ChatGPT Sites.

Single most important open question: Does the tool actually support educators in making better design decisions without over-relying on AI or erasing professional judgment?

Analysis basis: This report is based entirely on the self-reported, unverified description provided by the author. No third-party verification, revenue data, customer feedback or traction metrics are available.

The description states that Visible Thinking Designer is an AI-assisted design partner for tertiary and vocational educators. It allows tutors to bring one real learning activity, task, or assessment—either described directly or via PDF, Word document, or photograph—and then helps them:

• Clarify the intended capability;
• Consider what learners need to Know, Do, and Be & Relate;
• Identify technical, language, literacy, numeracy, and cultural demands;
• Establish a focused understanding of task, readiness, and AI’s role;
• Design 3–5 moments where consequential learner thinking could become visible;
• Produce an editable Visible Thinking Plan.

The tool does not use learner surveillance or automatic capability judgements. It supports educator professional judgement rather than replacing it.

Claim: The product is a web-based application built with Next.js, TypeScript, and OpenAI APIs.

Evidence: The author states they used “ChatGPT Sites”, “Next.js”, “OpenAI JavaScript SDK”, “GPT-5.6”, and “structured-outputs” to build it.

The author positions Visible Thinking Designer as a tool that addresses the shift in education where "answers become abundant" but evidence of learning is harder to interpret. It focuses on making learner thinking visible—not just polished outputs—as evidence of capability.

It is described as emerging from fieldwork with vocational educators in Aotearoa New Zealand, rooted in educational research and grounded in a model tested through practice.

Claim: The tool is designed to help educators avoid false certainty when using AI.

Evidence: The author says: “The central challenge was avoiding false certainty. Visible Thinking Designer must help educators make better design decisions without pretending that a language model can determine whether learning has occurred.”

The target customer is tertiary and vocational educators in Aotearoa New Zealand, specifically those working with adult literacy and numeracy assessments.

Claim: The tool targets tutors who want to redesign real tasks so that learner thinking becomes visible.

Evidence: The description states: “Visible Thinking Designer is an AI-assisted design partner for tertiary and vocational educators.” And: “Tutors could see important learner thinking becoming visible through attempts, questions, practical decisions and professional conversations... but this evidence was often incidental, unrecognised or missing from the task and assessment design.”

Not evidenced.

Claim: No business model or pricing information is provided.

Evidence: The description does not mention any revenue streams, pricing tiers, subscriptions, or monetization strategy.

The tool is built using:

• Framework: Next.js
• Language: TypeScript
• AI Integration: OpenAI JavaScript SDK, GPT-5.6, Responses API
• Validation: Zod schemas
• Interface rendering: Deterministic application code
• File handling: Supports PDFs, Word docs, images; processes attachments via OpenAI file/image inputs
• Persistence: Browser-local storage (no account required)
• Testing: 62 automated tests, TypeScript checking, linting, production builds

Claim: The tool uses structured prompts and deterministic rendering.

Evidence: The author says: “The model interaction is divided into three versioned prompt contracts... Strict Zod schemas validate requests and structured model outputs. Deterministic application code then renders the final plan rather than asking the model to generate an uncontrolled document.”

Not evidenced.

Claim: No data on usage, adoption, or traction is provided.

Evidence: The description states that this is a public prototype and that testing with educators is planned. There are no mentions of users, customers, or real-world deployment beyond the author’s own use cases.

Not evidenced.

Claim: No mention of competitors or market positioning.

Evidence: The description does not reference existing tools in the educational design space or AI-assisted learning platforms.

• Over-reliance on AI without clear boundaries: While the tool claims to support professional judgment, it is built around AI prompting and structured outputs. Risk of over-dependence if not carefully managed.

• Limited scope and maturity: The MVP is described as focused on one task per educator, with no plans for broader platform features or scalability.

• Unverified impact claims: The author makes strong claims about educational outcomes but offers no data to back them up.

Inference: If the tool becomes widely adopted, there may be risks around how it shapes pedagogical practices and whether it reinforces certain assumptions about learning.

1. How do you validate that the tool actually improves educational design decisions?
2. What specific feedback have you received from educators during testing?
3. Are there any plans to collect or analyze usage data, even if not currently stored?
4. How do you ensure that AI does not silently override educator judgment in practice?
5. What is your long-term vision for scaling beyond the current prototype?