Skip to content

docs: add interactive interview preparation guide - #17

Closed
Akaliup wants to merge 1 commit into
masterfrom
feat/rag-hybrid-retrieval
Closed

Akaliup wants to merge 1 commit into
masterfrom
feat/rag-hybrid-retrieval

Conversation

@Akaliup

@Akaliup Akaliup commented Nov 11, 2026

Copy link
Copy Markdown
Owner

变更摘要

新增交互式面试准备指南,整理 51 个基于项目实现、评测结果和已知限制的问题,覆盖业务口径、架构、LLM、RAG、审批、安全与评测。页面提供搜索、分类筛选、掌握进度、主题切换、展开答案和打印功能,帮助按模块准备项目介绍与技术追问。

关联任务

无

实际验证

  • python -m pytest backend/tests -q
  • npm --prefix frontend test
  • npm --prefix frontend run build
  • python scripts/smoke.py(涉及 HTTP/前端流程时)
  • docker compose config --quiet(涉及 Compose 时)
  • 真实 LLM 调用(如执行,注明次数、模型和报告路径;自动 CI 不执行)
  • 未执行项已在下文明确标注

验证结果:

未执行自动化验证;本次变更仅新增静态 HTML 文档。

安全与边界

  • 未提交 .env、API Key、数据库、缓存、构建产物或本机运行数据
  • 所有执行仍为 simulation,没有真实广告、采购、支付或外部系统副作用
  • 金额仍以 USD 整数美分存储和传输
  • 服务端仍决定角色,未信任客户端角色请求头
  • 审批与执行仍依赖条件更新、事务和唯一约束
  • 未在自动 CI 中调用真实 LLM

契约与文档

  • 契约未变更,无需同步 docs/CONTRACT.md、后端测试或前端类型
  • 未涉及评测逻辑变更,无需更新 docs/EVALUATION.md
  • 指南说明了演示认证、单 worker、合成 SOP 和模拟执行等项目限制

审查提示

请重点核对指南中的实现细节、评测数字及限制是否与仓库当前状态一致,并检查金额守恒、证据 ID 可追溯性、LLM 失败/降级可见性、审批执行幂等和敏感信息。自动审查工具的评论只是辅助,不替代人工审查。

- Add 51 repository-grounded interview questions with follow-up prompts
- Include search, filters, progress tracking, theme toggle, and print support
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Nov 11, 2026 •

Copy link
Copy Markdown
Codex Review Summary: ✅ Completed

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-11-11T15:21:52.305579Z 24dbdf6 PR opened
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@Akaliup Akaliup closed this Nov 11, 2026

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 24dbdf6f0e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread docs/interview-prep.html
<div class="answer">
<ul>
<li>scrypt,<code>N=2^14</code>、<code>r=8</code>、<code>p=1</code>,随机 salt;constant-time 比较;不存明文/可逆加密。</li>
<li>登录限流 10/min;身份取 accepted token/JWT;无效 token 归入匿名桶,防止攻击者用随机 token 绕过。</li>

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Correct the login rate-limit identity

For /api/auth/login, request_context in backend/app/main.py explicitly clears the bearer token and uses the shared per-host anonymous bucket; accepted token/JWT identities apply to other scopes. This bullet says the opposite and contradicts the trap below it, so readers may repeat an incorrect security design. Distinguish login’s bucket from authenticated buckets on other routes.

Useful? React with 👍 / 👎.

Comment thread docs/interview-prep.html
<summary>参考答案 · 3 个追问</summary>
<div class="answer">
<ul>
<li>当前 6 篇 SOP、32 个 chunk;全表打分 P50 6.8ms(hashing),加向量库只增加运维复杂度。</li>

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Update the retrieval latency figure

This presents 6.8 ms as the current full-scan P50, but the current RAG report’s hashing results are 7.3 ms for weighted and 10.0 ms for hybrid_rrf; 6.805 ms is from the earlier baseline report. Label the older measurement with its pipeline/date or use the current benchmark so readers do not quote a stale figure.

Useful? React with 👍 / 👎.

Comment thread docs/interview-prep.html
function syncTheme() {
theme.textContent = root.dataset.theme === 'dark' ? '浅色' : '深色';
}
const savedTheme = localStorage.getItem('profitops-theme');

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Guard the theme storage read

If browser policy blocks storage or this page runs in a context where localStorage access throws, this unguarded read aborts the initializer. The final initial filter/progress update and later print, keyboard, copy, and navigation setup then never run, even though other storage accesses are guarded. Handle this read failure as the progress read does.

Useful? React with 👍 / 👎.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant