Skip to content
You must be logged in to sponsor guriguri215-lang

Become a sponsor to T Yamada

@guriguri215-lang

T Yamada

guriguri215-lang

I build and maintain local-first open-source research tools and prototypes for reproducible analysis. The work spans specification and evidence workflows, bounded game-theory models, retrospective poker analysis, and cost-aware bug-investigation experiments.

I publish code, documentation, tests, examples, and development notes so others can inspect assumptions, reproduce the work, and understand its limits. Where AI assistance is used, its output is treated as a proposal—not verified evidence—until applicable checks and evidence support it.

Sponsorship helps me spend more time on maintenance, testing and validation, documentation, reproducible examples, benchmark and evaluation work, and release preparation across these projects. It is optional support for the public work as a whole, not a purchase of features, priority support, custom work, consulting, poker advice, solver output, or guaranteed results.

Featured work

  1. guriguri215-lang/repeated-poker-analysis

    Research prototype for exact-response and bounded commitment analysis in small two- and three-player poker models.

    Python
  2. guriguri215-lang/bug-cause-inference-game

    Reproducible research prototype for cost-aware Bayesian bug investigation on synthetic, injected-bug, and fixed hand-authored toy benchmarks; not production fault localization.

    Python
  3. guriguri215-lang/poker-deliberation-framework

    Experimental local-first Python toolkit for auditable poker calculations and bounded retrospective NLHE river reviews; not real-time decision support or a full solver.

    Python
  4. guriguri215-lang/spec-driven-agent-framework

    Experimental offline-first Python CLI for validated specifications, context snapshots, host-execution intents, and bounded finite-domain solving; it does not call LLMs or launch agents.

    Python
  5. guriguri215-lang/DPL_poker_ai

    Early-alpha, simulation-only Python research framework for poker decision provenance, bounded CFR/CFR+ river experiments, synthetic leaks, and explanation-faithfulness checks; not a real-time bot.

    Python
  6. guriguri215-lang/decision-assurance-framework

    v0.1 source-only milestone: local-first Python framework for policy-gated decision support, evidence-linked reports, and explicit human handoff; advice-only, not autonomous.

    Python

Learn more about sponsoring developers and organizations.

Report abuse

Select a tier

$ a month

You'll receive any rewards listed in the $5 monthly tier. Additionally, a Public Sponsor achievement will be added to your profile.

$5 a month

Select

Optional support for maintaining open-source research tools, testing and validation, documentation, reproducible examples, benchmark and evaluation work, and release preparation.

This supports the public work as a whole. It does not include feature priority, custom implementation, consulting, poker advice, solver output, support hours, or guaranteed results.

No additional rewards or deliverables are promised.