10 RAG Projects That Take You From Basic Retrieval to Production-Grade AI Systems
_jaydeepkarale · x · 2026-10-07
Developer jaydeepkarale outlines 10 projects that go beyond the classic "PDF chatbot," covering the full path from retrieval basics to production systems:
- Hybrid Search RAG: combine BM25 keyword search with dense vector retrieval
- Metadata-Filtered RAG: semantic search plus date/category filters
- Reranking RAG: retrieve many candidates, rerank with a CrossEncoder, send only top chunks to the LLM
- Contextual Chunking RAG: preserve document-level context when chunking
- SQL + Vector RAG: mix structured queries with unstructured knowledge
- Knowledge Graph + Vector RAG: add entity relationships via Neo4j
- Corrective RAG: rewrite queries and re-retrieve when context is weak
- Self-RAG: let the system decide when to retrieve and verify answer support
- Multimodal RAG: retrieve and reason over text, tables, images, and document pages
- Agentic RAG: an agent picks knowledge sources and performs multi-step retrieval
The key insight: these architectures are composable — hybrid search + reranking + metadata filtering + corrective retrieval can merge into one production RAG pipeline aiming for accurate, contextual, verifiable, adaptive retrieval.
More from coding & agent
- One-line AGENTS.md tweak: tell your coding agent you're a tired engineer — cem2ran · 2026-10-07
- mitsuhiko: the bash tool makes codemode surprisingly hard to explain — mitsuhiko · 2026-10-07
- Agent harness replicates Minecraft builds: maps, places blocks and generates its own assets — Promptmethus · 2026-10-07
- Developer has Codex build a Solarman solar-monitoring macOS app in 3 minutes — Dimillian · 2026-10-07
- Open-source Claude skill huashu-art-motion: AI-made science video hits 100K views on Douyin in 5 hours — AlchainHust · 2026-10-07
- Redditor passes three Claude certification exams and shares his full study playbook — Few-Association-913 · 2026-10-07