Section

Cases

Case studies of notable hackathons — both exemplary structures and instructive failures.

Cases

Stub. Drafted during Phase 1 IA skeleton seeding. The piece below is a one-paragraph placeholder — replace before publishing.

Case studies of notable hackathons — both exemplary structures and instructive failures.

All studies (16)

  • The canonical Blind Replication Sprint. The four-team blind imaging protocol, the October 2017 EHT Image Hackathon, the seven-week 2018 sprint, the July 2018 convergence workshop reveal, and the 2019 Bouman credit-attribution episode as a named storytelling failure mode.

  • The exemplary game jam. Theme plus 48 hours, peer judging dominant, genre conformity as a named failure mode.

  • The canonical Platform Ecosystem Challenge. Includes the structural shift from a single global stage to regional competitions in 2025. Participant-authored problem statements with the Impact-plus-Technology rubric.

  • Hackathon-to-acquisition. The narrative arc that made GroupMe legible to acquirers in the demo window.

  • The long-running historical Platform Ecosystem Challenge that predates the modern AI era. What it carried forward and what it stopped doing.

  • The original code sprint, since 1999. Single-codebase, non-competitive, conference-adjacent. Defines the archetype.

  • Founder-track / Startup Weekend lineage in the AI era. Pitch-judged, business-track, ringer-tolerant on purpose.

  • The internal corporate hackathon at scale. Guinness World Record event documenting the AI-era shift in what "competent participant" means.

  • ETHGlobalDraft

    The sponsor-bounty federation archetype. Partner-prize architecture, async-plus-live two-round judging, AI-attribution rules that have become quietly industry-shaping.

  • A themed multi-track that picked a frame (civic impact) and committed. What changed in the team dynamics.

  • HackMITDraft

    University season hackathon. Science-fair judging at scale, novice-friendly framing, the recruiter dynamic.

  • The exemplary themed multi-track. Four-tier judging cascade, five-criterion rubric, ten award categories, SME-authored Challenge Statements. Self-described as collaborative, not competitive.

  • PennAppsDraft

    University season hackathon, hardware-leaning. Demo-stage and hardware-stage bias as named failure modes.

  • The canonical ringer scandal. What happens when a fair-fight event fails to label its eligibility rules clearly.

  • Government civic challenge at national scale. Per-Ministry problem-statement model, college-internal pre-screening, software-versus-hardware split.

  • TreeHacksDraft

    Stanford's university season hackathon. Track-balance trade-offs and the science-fair format.