skills.rest
SoubhagyaJain
SoubhagyaJainCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

production-rag

Design production RAG systems with hybrid retrieval, reranking, and citation-grounded generation.

1|Updated Jun 17, 2026
Search Tags:#evaluation#citations#chunking#reranking#ollama#hybrid retrieval#production rag
One-click install
npx skills add https://github.com/SoubhagyaJain/Rag-chatbot --skill production-rag
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: production-rag
Source: https://github.com/SoubhagyaJain/Rag-chatbot/tree/main
Command: npx skills add https://github.com/SoubhagyaJain/Rag-chatbot --skill production-rag

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you design and improve retrieval-augmented generation systems that stay grounded, measurable, and dependable in production instead of collapsing into hallucinations, weak retrieval, or opaque failures.

Core Features & Use Cases

  • Production RAG Architecture: Plan end-to-end retrieval, reranking, grounding, citation, and fallback behavior for serious document assistants.
  • Local and Hybrid Stacks: Apply it to Ollama, ChromaDB, LlamaIndex, Streamlit, and other self-hosted or privacy-sensitive deployments.
  • Evaluation and Reliability: Build golden sets, track faithfulness and relevancy, tune chunking and hybrid retrieval, and add guards for code-heavy or compliance-sensitive answers.
  • Use Case: Use this skill when you need a company-policy assistant that returns cited answers, abstains when evidence is missing, and supports iterative improvement after every pipeline change.

Quick Start

Ask me to design or debug a production RAG system for your corpus, and I will give you a grounded architecture, retrieval strategy, evaluation plan, and deployment recommendations.

Frequently Asked Questions about production-rag

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a production RAG system that prevents hallucinations and grounds answers in citations?▼

A production RAG system prevents hallucinations by combining hierarchical chunking, citation-grounded generation, and faithfulness guards. This Skill designs end-to-end retrieval pipelines that enforce abstention when evidence is missing, ensuring dependable and grounded outputs.

What's the best way to set up hybrid dense and BM25 retrieval for a local Ollama and Chroma stack?▼

Setting up hybrid retrieval for local Ollama and Chroma stacks involves combining dense vector search with BM25 algorithms and applying reranking. This Skill designs these retrieval strategies to optimize document grounding for privacy-sensitive, self-hosted deployments.

How do I evaluate retrieval-augmented generation reliability and track faithfulness for compliance-sensitive corpora?▼

Evaluating retrieval-augmented generation reliability requires building golden sets and tracking faithfulness and relevancy metrics. This Skill provides a repeatable measurement framework to tune chunking and monitor pipeline quality after every change for compliance-sensitive corpora.

Does this approach support agentic workflows and metadata filtering for dynamic corpora?▼

This approach supports agentic workflows and metadata filtering for dynamic corpora. The Skill designs production-ready architectures incorporating fallback behavior and robust metadata filtering to handle evolving, compliance-sensitive document sets effectively.

Why does my RAG pipeline return weak retrieval results and how do I fix it?▼

Weak retrieval results in a RAG pipeline often stem from improper chunking or missing reranking layers. This Skill helps debug by tuning hierarchical chunking, implementing hybrid dense-plus-BM25 retrieval, and applying reranking to improve context extraction.

Related Skills

Explore More Software Engineering Skills→
hmzainjamilhmzainjamil
1

data-extractor

Extract structured text, tables, and metadata from documents into machine-readable output.

Community
Advanced
MantisWareMantisWare

eval-rag

Evaluate RAG pipelines with retrieval and generation metrics and bottleneck analysis.

Community
Advanced
constructive-ioconstructive-io

pgvector-rag

Create and manage pgvector databases in PostgreSQL with Ollama for RAG pipelines.

Official
Advanced
rizaldiemrizaldiem

rag

Build retrieval-augmented generation pipelines with vector databases and document chunking.

Community
Advanced