The discovery layer for software engineering knowledge. Search, browse, and explore across engineering blogs, YC startups, Product Hunt launches, GitHub repos, Hacker News, Stack Overflow, npm packages, Dev.to, Medium, Reddit, Wikipedia, and our own curated knowledge curriculum — all in one place.
Software engineering knowledge is scattered across thousands of platforms — blogs, forums, package registries, startup directories, documentation sites, and educational resources. Each platform has its own search, its own UX, its own silo.
100xSystems breaks those silos.
We are not a content aggregator. We are a search + discovery registry layer — a single entry point that indexes metadata from across the web and presents it in a unified, beautiful, fast interface. We don't store your content. We surface it, connect it, and help engineers discover what matters.
- Registry-first, not storage-first. We index metadata and links. The content lives on the original creator's platform — we just help people find it.
- Curated knowledge curriculum. Our knowledge graph (principles, languages, tools, patterns) provides structured, interconnected engineering concepts sourced from Wikipedia + Wikidata — built by humans, not AI slop.
- Multi-source search. One query spans local cached data (feeds, YC companies, PH products, knowledge graph) and live web APIs (GitHub, HN, SO, npm, Reddit, Medium, Dev.to, DDG, Wikipedia — and growing).
- Open source, non-commercial. All data is sourced from Creative Commons, public APIs, and open repositories. We give credit and attribution. This is an open source educational platform.
- Scalable by design. From 100 feeds to 10,000. From 5 sources to 500. The architecture is designed to grow horizontally — more JSON files, more API sources, more categories.
- Knowledge expansions: MDN Web Docs, JavaScript.info, freeCodeCamp, and other Creative Commons resources — indexed and cross-linked at the concept level.
- Resource packs: Illustration collections, design assets, code snippets — with a single "download all" button that scrapes and bundles from their original sources.
- Custom curricula: User-defined learning paths across concepts, tools, and patterns.
- Offline catalog: A downloadable registry of all indexed content for air-gapped study.
- Type-specific pages: Separate routes for YC companies, Product Hunt launches, GitHub trending, news feeds — each with their own filtering, sorting, and exploration UX.
100xsystems/
├── registry/ ← Central index (JSON metadata files)
│ ├── feeds/ ← RSS/Atom feed data (daily delta)
│ ├── knowledge/ ← Curated knowledge curriculum (principles, languages, tools, patterns)
│ │ ├── principles/
│ │ ├── languages/
│ │ ├── tools/
│ │ ├── patterns/
│ │ ├── manifest.json
│ │ └── seeds.json
│ ├── yc/ ← Y Combinator company catalog (Algolia-sourced)
│ ├── producthunt/ ← Product Hunt product catalog
│ └── scripts/ ← Workflows to update data
│ ├── github-workflow/ ← Daily automation scripts
│ └── one-time/ ← Manual seed scripts
│
└── website/ ← Next.js static site (this repo)
├── app/ ← Pages & API routes
│ └── api/search ← Multi-source live search API
├── src/
│ ├── presentation/features/ ← Homepage sections & components
│ └── knowledge/ ← Knowledge graph types
└── public/
├── feed-cache.json ← Latest feed items (cloned from registry)
├── yc-cache/ ← YC companies (cloned from registry)
├── ph-cache/ ← Product Hunt products (cloned from registry)
└── knowledge-cache/ ← Knowledge graph (cloned from registry)
Registry (GitHub)
│ Daily workflow updates feeds, YC changes, PH daily
▼
Website build / dev
│ Clones registry → caches JSON files → generates static content
▼
Next.js SSG
│ Pre-renders pages with cached data at build time
▼
Client browser
│ Homepage: loads cached JSON + live API search
▼
User
← Unified search across local + live sources
The search works in two layers:
-
Local (SSG) layer — Data cached at build time (feeds, YC companies, PH products, knowledge graph). Searched via Fuse.js (lightweight fuzzy search). Zero latency, works offline, instant results.
-
Live API layer — Real-time queries to external APIs (GitHub, HN, SO, npm, Reddit, Medium, Dev.to, DDG, Wikipedia). Each source is fetched in parallel with per-source error isolation.
The results are merged client-side into a unified, sectioned interface with unique card components per source type, brand favicons, source-specific metadata, and a single-select filter.
# Install dependencies
npm install
# Run development server (clones latest registry data)
npm run dev
# → Starts at http://localhost:3000
# Production build
npm run build
# → Generates static site with fresh cacheThe dev script (npm run dev) automatically clones the latest registry data before
starting the server, so you always work with fresh data.
| Source | Description | Update Cadence |
|---|---|---|
| Engineering Blogs | 400+ RSS/Atom feeds from top engineering blogs | Daily via registry workflow |
| YC Companies | 6,000+ Y Combinator companies (full catalog + daily changes) | Daily via Algolia API |
| Product Hunt | 900+ products (full catalog + daily trending) | Daily via PH API |
| Knowledge Graph | 162 curated engineering concepts (principles, languages, tools, patterns) | Manual curation |
| Source | API | Free Tier |
|---|---|---|
| Hacker News | Algolia HN Search | Unlimited |
| GitHub | GitHub REST API | 5,000 req/hr with PAT |
| Stack Overflow | Stack Exchange API | 300 req/day |
| npm | npms.io API | Unlimited |
| Dev.to | DEV.to API | Unlimited |
| Medium | RSS feeds per tag | Unlimited |
| DuckDuckGo | Instant Answer API | Unlimited |
| Reddit HTML parsing | Rate-limited | |
| Wikipedia | Wikipedia REST API | Unlimited |
- W3Schools
- MDN Web Docs
- JavaScript.info
- freeCodeCamp
- Creative Commons resource directories
- Illustration & design asset catalogs
The knowledge curriculum organizes 162 software engineering concepts into 4 categories:
principles/ (34) — ACID, CAP Theorem, SOLID, DRY, Circuit Breaker, etc.
languages/ (28) — Rust, Go, TypeScript, Python, WebAssembly, etc.
tools/ (56) — Kubernetes, Docker, React, PostgreSQL, Kafka, etc.
patterns/ (44) — Bloom Filter, CQRS, Event Sourcing, Saga, etc.
Each entity has:
- Wikipedia description + summary
- Wikidata references
- External URLs
- Parent/child/related relationships (future: full graph traversal)
The knowledge section appears first in search results — before any other source — because it represents curated, structured engineering knowledge.
- Unified search across local + live sources
- Brand SVGs and favicons for every source
- Unique card components per source type
- Single-select source filter
- 1-second debounce
- Daily registry workflow
- Knowledge curriculum as first search section
- Wikipedia API integration
- Fix source pill navigation (no more query prefix pollution)
- Meaningful homepage hero heading
- Separate route pages per source type
- MDN Web Docs integration
- JavaScript.info / freeCodeCamp indexing
- W3Schools reference integration
- Creative Commons content registry
- Dark mode
- Share/bookmark per result
- "Download all" for resource packs
- User-curated collections
- 5,000+ feed sources
- 50+ live API sources
- Full knowledge graph traversal (parent/child/related navigation)
- Semantic search (alternative to keyword-only)
We use data from the following platforms under their respective terms. All content remains on the original creator's platform. We index metadata and provide links:
- Wikipedia — Creative Commons Attribution-ShareAlike
- YC Combinator — Public Algolia API / public directory
- Product Hunt — Official v2 API
- GitHub — Public REST API
- Hacker News — Algolia Search API
- Stack Overflow / Stack Exchange — CC BY-SA, Stack Exchange API
- Dev.to — Forem API
- Medium — RSS feeds
- DuckDuckGo — Instant Answer API
- npm — npms.io API
- Reddit — Public data (old.reddit.com HTML)
MIT — see LICENSE.
Built with ❤️ by the 100xSystems team.