smile-struggler

smile-struggler / bagu-killer

Public

八股杀手，专杀八股！基于大模型的大厂面经八股采集器，一键完成采集到筛选到成文全流程，支持自定义关键词，让所有八股都无处遁形！

48
6
89% credibility
Found Apr 27, 2026 at 48 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A tool that collects AI Agent interview experiences from social media posts, uses AI to extract and deduplicate questions, and generates organized Markdown guides for job preparation.

How It Works

1
🔍 Discover Bagu Killer

You stumble upon this helpful tool on GitHub that gathers real people's AI job interview stories and turns them into a tidy question list.

2
💻 Set up your collector

Download the tool to your computer and get it ready by linking a simple storage spot for your growing collection.

3
🕸️ Start gathering stories

Point the tool at social media posts sharing genuine AI Agent interview experiences to begin collecting them.

4
🤖 AI processes the magic

Watch as smart helpers scan texts and pictures, filter out ads, pull out questions, and group similar ones together neatly.

5
🔄 Refresh daily

Run a quick update each day to snag the newest interview tales and keep your list fresh.

6
📚 Explore your question bank

Open beautiful guides listing top questions by type, with counts from companies and roles.

🎉 Nail your interviews

Dive into high-frequency questions from real stories, feeling confident and ready for your AI job adventure.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 48 to 48 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is bagu-killer?

This Python tool automates scraping Xiaohongshu for real-world AI Agent interview experiences from big Chinese tech firms, using LLMs for classification, image OCR, question extraction, and semantic deduplication into a canonical "bagu" question bank. Run one CLI command like `ai-offer pipeline daily-sync` with custom keywords, and it handles crawling, filtering fakes/ads, merging similar questions via embeddings, then spits out Markdown reports with high-frequency questions grouped by type (knowledge QA, LeetCode, projects). Devs get a traceable, stats-rich prep resource showing what gets asked across companies and roles, no manual copying needed.

Why is it gaining traction?

Unlike raw face经合集 or static lists, it builds dynamic, image-inclusive databases that prioritize repeated "拷打" questions from verified posts, with post counts, company ties, and role breakdowns. The daily sync script and JSON/CSV report exports make it dead simple to maintain fresh data, while custom keywords target niches like "tool calling" or "RAG." Python CLI users love the end-to-end pipeline that turns noisy social data into actionable intel without babysitting.

Who should use this?

AI engineers prepping for ByteDance, Alibaba, or Tencent Agent/ML interviews, especially those grinding high-frequency bagu on LeetCode, projects, or LLM internals. Chinese devs sifting Xiaohongshu for authentic 凉经/面经, avoiding paid course spam. Teams building internal question banks for mock interviews.

Verdict

Grab it if you're in the Chinese AI job hunt—solid CLI and auto-generated docs make setup fast, despite 48 stars signaling early maturity. Credibility score of 0.8999999761581421% flags it as experimental; test on your hardware with Qwen models before production reliance.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.