Prompt Problems: A New Programming Exercise for the Generative AI Era

This is an academic paper introducing a new pedagogical framework for the LLM era rather than a software project. With 0 stars, 0 commits/hour velocity, and 879 days of age, this is a dormant research publication with minimal adoption or engineering follow-through. The work is conceptually interesting—reframing programming education around prompt engineering and LLM output evaluation is timely—but it lacks any evidence of community traction, maintained code, or production deployment. The 7 forks suggest academic citations or interest but no active development. As a reference implementation, it would serve primarily as inspiration for educators or researchers building their own exercise platforms. Defensibility is extremely low because: (1) the core contribution is pedagogical and conceptual, not a defensible technical artifact; (2) major platforms (OpenAI, Google, Anthropic) are already building educational content around LLM usage and prompt engineering; (3) any institution could easily design their own prompt-based programming exercises without adopting this framework. Platform domination risk is high because OpenAI's curriculum initiatives, Google's AI-focused education programs, and emerging EdTech platforms are rapidly absorbing this exact domain—teaching prompt engineering as a first-class skill. Market consolidation risk is medium because while no single EdTech incumbent has locked in this niche, the barrier to entry is low and capital-backed competitors could easily outspend academic development. Displacement horizon is 1-2 years as mainstream educational platforms begin shipping prompt-based coding challenges as native features. The paper likely contributes genuinely novel framing to computing education but lacks the implementation depth, adoption, or technical moat to defend against commercial or platform-driven alternatives.