一份合同下多个项目、一年内按名册批量：逐项目跑完整交付流，协调人视图一屏看到全批的合规风险、逾期项与批量倒排。面向独立教练、工作室、学校与机构。One contract, multiple projects, run down a roster across a year: the full delivery flow per project, with a coordinator view showing the whole cohort's compliance risk, overdue items, and batch countdown on one screen. Built for independent coaches, studios, schools, and institutions.
ISEF 没有禁止培训机构或营利机构指导参赛项目的规则，机构学校版就是明确面向独立教练、工作室、学校与机构设计的：一份合同覆盖名下多个 ISEF 项目，一年内按名册批量运行。我们不担任任何官方角色（Adult Sponsor、Qualified Scientist、Designated Supervisor 都不是我们），只做审计、提问与跟踪。ISEF has no rule prohibiting mentoring by training or for-profit institutions, and the Institution tier is built explicitly for independent coaches, studios, schools, and institutions: one contract covering several ISEF projects, run down a roster across a year. We take on no official role — we are not the Adult Sponsor, Qualified Scientist, or Designated Supervisor — we audit, question, and track.
带一批项目时最难的不是判断单个项目好不好，而是不同项目踩在不同的时序上：一个还没拿到 IRB 裁定就开始了招募，一个原型已经在成人身上试用过但没做风险评估，一个是延续项目却没区分往年与本年工作，一个只是 AI 使用日志没整理。这四类风险的责任机构、最晚日期和负责人完全不同，靠记忆管不住。The hard part of running a cohort is not judging whether one project is good — it is that different projects sit on different timing failures: one started recruiting before its IRB determination, one had a prototype tested on adults with no risk assessment, one is a continuation project that never separated prior from current work, one merely has an untidied AI-use log. Those four carry different responsible bodies, different latest dates, and different owners; memory does not hold them.
机构学校版给的是这张表：全批横向风险总表、逾期与无进度项、按各自 D-0 对齐的批量倒排，以及每个项目自己的 M1 / M2 / M3。单个项目被判 blocked 不中断队列：记为 blocked 并附原因，继续下一个。只带一个项目的，请看 学生赛季版。The Institution tier hands you that table: a cross-cohort risk summary, overdue and stalled items, a batch countdown aligned to each project's own D-0, and each project's own M1 / M2 / M3. A single blocked project does not stop the queue — it is recorded as blocked with the reason and the queue moves on. Running just one project? See the Student tier.
该选的情况：工作对象已经从"一个项目怎么补"变成"一批项目怎么排、怎么看、怎么追"；需要横向风险总表和按各自截止对齐的倒排；项目成熟度参差，希望卡住的被看见、其余的继续走。Choose this tier when the job has shifted from “how do we fix one project” to “how do we order, watch and chase a batch”: you need the cross-project risk table, countdowns aligned to each project’s own deadline, and a queue where a stuck project gets flagged while the rest keep moving.
只有一个学生、一个项目，也不需要名册和汇总表：学生赛季版更合适，甚至免费套件就够，别为用不上的队列付费。One student, one project, no roster or summary table needed? Student is the better fit — the free suite may even be enough. Don’t pay for a queue you won’t use.
想要跨机构对比、排名或“哪个项目更有希望”的排序：我们不做这类输出。协调人视图列的是状态和风险，不是优劣。Cross-institution comparisons, rankings, or a “which project has the best shot” ordering? We don’t produce that output. The coordinator view lists status and risk — never merit.
名册上的每个项目都跑完整的 Student 交付流；在此之上加名册管理、批量队列与协调人 / 政策视图。全部交付件出 HTML / PDF / DOCX 三格式，可用机构品牌。Every project on the roster runs the full Student delivery flow; on top of that come roster management, the batch queue, and the coordinator and policy views. All deliverables ship in HTML, PDF, and DOCX, and institutional branding is available.
blocked 并附原因，队列继续With no substantive student draft, M1 does not start; the project is recorded as blocked with the reason and the queue continuescohort_id / project_id），代号绝不使用学生姓名或其变体；名册只含非敏感字段The roster uses code names (cohort_id / project_id), and a code name never uses a student's name or a variant of it; the roster holds non-sensitive fields only<cohort_id>/<project_id>/ 分目录，交付件命名统一为「项目代号-交付件名-日期」Archiving is nested under <cohort_id>/<project_id>/, with deliverables named "project code - deliverable name - date"以及不论哪个档位、花多少钱都不提供的：代写研究计划 / 摘要 / 海报文本 / 论文初稿；晋级或获奖预测。这两条是红线，不是待解锁的高级功能。名册与报告中不出现学生姓名、出生日期、联系方式、参与者身份或知情同意书正文。And what no tier has at any price: ghostwritten research plans / abstracts / poster text / paper drafts, and advancement or award prediction. These are red lines, not locked premium features. Rosters and reports contain no student names, dates of birth, contact details, participant identities, or informed-consent text.
机构学校版首轮免费：先按名册把交付做完，贵方先看交付内容，觉得有用再谈合同。合同价格在试点之后逐单回信确认。The first round of the Institution tier is free: we run the roster and you judge the work first. Contract pricing follows the pilot, confirmed per order by email.
批量交付的排期逐单在回信里约定。事前审批的时序是官方规则的硬约束，不是我们能压缩的。Batch scheduling is agreed per order by email. Prior-approval timing is an official hard constraint — not something we can compress.
不想写邮件的话，填几项基本信息就行，我们会主动联系你。No email needed — a few basics and we’ll reach out to you.
能。ISEF 没有禁止培训机构或营利机构指导参赛项目的规则，机构学校版就是为独立教练、工作室、学校与机构设计的。我们唯一会婉拒的两类请求是：要求代写学生材料，以及要求预测晋级或获奖。这两条与你是不是机构无关，对家长、学生、教练一视同仁。Yes. ISEF has no rule prohibiting mentoring by training or for-profit institutions, and the Institution tier is designed for independent coaches, studios, schools, and institutions. The only two requests we decline are ghostwriting student materials and predicting advancement or awards — and those apply regardless of whether you are an institution, equally to parents, students, and coaches.
不会。单个项目失败（例如缺学生初稿、缺事前审批证据）记为 blocked 并附原因，队列继续跑下一个；协调人视图会把它单列在逾期与无进度表里，并给出今日动作。No. A failed project (missing student draft, missing prior-approval evidence) is recorded as blocked with the reason and the queue moves on; the coordinator view lists it separately in the overdue-and-stalled table with an action for today.
不会。名册与所有交付件一律用项目代号，代号不使用学生姓名或其变体；学生姓名、出生日期、联系方式、参与者身份、知情同意书正文与任何医疗标识符都不进入台账、名册、报告正文或文件名。官方表单原件由学校或责任机构保存，我们只记录最小状态与非敏感证据指针。No. The roster and every deliverable use project code names, and a code name never uses a student's name or a variant of it; student names, dates of birth, contact details, participant identities, informed-consent text, and any medical identifier never enter the ledger, roster, report bodies, or filenames. Official form originals stay with the school or responsible institution — we record only minimal status and non-sensitive evidence pointers.
因为它是单独计价的人工工作量，而不是档位自带的能力。机构学校版可以按合同配额购买：配额内的项目按已购处理，配额外的项目在协调人报告中如实标注"未含人工复核"，我们不会含糊过去。Because it is separately priced human effort, not a capability the tier carries by itself. The Institution tier can buy it as a contract quota: projects inside the quota are treated as purchased, and projects outside it are honestly marked "no human review included" in the coordinator report — we do not blur that line.
不能。任何档位都不输出晋级或获奖概率、名次预测，也不用与往届项目的相似度当作胜率代理。这是产品的红线，不是内测阶段的限制。我们能给的是证据与时序层面的判定：哪些项目卡在事前审批上、哪些主张没有数据支撑、哪些整改项已经逾期。No. No tier outputs advancement or award probabilities, placement forecasts, or similarity-to-past-winners as a win-rate proxy — that is a product red line, not a beta-stage limitation. What we can give is evidence- and timing-level verdicts: which projects are stuck on prior approvals, which claims lack supporting data, and which remediation items are already overdue.