← Reddit

Claude for teachers

Reddit · sariitha · July 3, 2026
A European primary school teacher uses Claude to design and organize lessons, finding it significantly superior to OpenAI for worksheet creation tasks. Although Claude struggles with generating images for educational materials, the teacher seeks strategies for efficiently producing thematic work bundles across different grade levels.

Detailed Analysis

A Reddit post from a European primary school teacher offers a grassroots glimpse into how educators are adapting Claude for classroom production work, specifically the creation of religion-class worksheets across multiple grade levels. The teacher describes building a fairly sophisticated personal toolkit around Claude—including a workflow markdown file, a "work bundle" skill, a design system document, and project-level instructions for use within Claude's collaborative "Cowork" environment—to streamline lesson planning and worksheet generation. This setup reflects a broader pattern among power users who go beyond simple prompting and instead construct semi-structured systems (documents, templates, reusable skills) that let Claude operate more like a customized production assistant than a one-off chatbot.

The post also surfaces a comparative claim that will interest both educators and AI vendors: the teacher states that Claude is "100 times better than OpenAI" for this specific use case of designing instructional worksheets, while simultaneously noting a clear weakness—Claude's inability to generate compelling images to make worksheets visually engaging for children. This bifurcation is telling. Anthropic has positioned Claude primarily as a text- and reasoning-oriented model, with image generation historically outsourced to or absent compared to competitors like OpenAI's DALL-E integration or Google's Gemini/Imagen tools. For a teacher-facing use case where visual appeal is pedagogically important (engaging young children with imagery), this gap is a meaningful limitation, and it underscores the tradeoffs users navigate when choosing between AI ecosystems: strong document/text reasoning versus multimodal content generation.

The specific request—help scaling personalized worksheet creation "in bulk" across different grade levels and rotating biweekly themes—points to a real operational challenge in education: differentiated instruction at scale. Teachers must produce varied materials tailored to different age groups and reading levels while maintaining thematic consistency, a task that is time-consuming without automation. Claude's strength in structured document generation, following detailed formatting and style instructions (as evidenced by the teacher's design system and workflow docs), makes it well-suited to templated, rules-based content production, which is likely why it outperforms alternatives here. This aligns with Anthropic's broader enterprise and education push, including Claude for Education initiatives and API/project features aimed at knowledge workers who need consistent, instruction-following behavior over creative flair.

More broadly, this anecdote is a microcosm of how AI adoption is unfolding in real-world professional settings: not through flashy demos, but through incremental, DIY infrastructure-building by individual practitioners who stitch together custom workflows, skills, and project instructions to solve mundane but high-value problems. It also highlights an unmet market need—education-specific tooling for visual content generation—that Anthropic, or third-party developers building on Claude's API, could address to capture the classroom-tools market more fully. As AI companies increasingly court education as a vertical (alongside coding, customer service, and enterprise knowledge work), gaps like image generation for worksheets represent both a current limitation and a clear signal of where future feature investment might be directed.

Read original article →