Hey devs! I want to share a project I have been working on called SearchWala (Swift-Search-RS).
What is it?
SearchWala is an open-source meta-search engine written in Rust that queries 90+ search engines simultaneously including Google, Bing, DuckDuckGo, Brave, Yandex, Baidu, and many more. It aggregates the results, scrapes the actual content, and can synthesize an AI-powered answer using any LLM provider you bring (Groq, OpenAI, Cerebras, etc).
Key Features
- 90+ Search Engines: Google (15 regional variants), Bing, DuckDuckGo, Brave, Yahoo, Qwant, Mojeek, Startpage, Wikipedia, and 60+ more
- Dual LLM Pipeline: Lite mode (fast BM25-ranked answer) and Research mode (iterative deep analysis across 200+ sources)
- 5-Tier Content Extraction: Intelligent article extraction that goes beyond simple scraping
- Built in Rust: Blazing fast with async Tokio runtime, 24 concurrent scrapers
- BYOK (Bring Your Own Key): Works with any OpenAI-compatible LLM provider
- Privacy First: No tracking, no ads, self-hostable
- Docker Ready: Multi-stage build producing a tiny 15MB image
Performance
- Scrapes and ranks 420 URLs in under 10 seconds
- BM25 ranking with phrase match bonuses
- Browser fingerprint rotation across 12 profiles to avoid blocks
Try It
GitHub: https://github.com/SandeepAi369/Swift-Search-Rs
Would love to hear your feedback! Drop a comment if you have questions.
Top comments (0)