skills.rest
thesammykins
thesammykinsCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

llm-rag-engineer

Build RAG pipelines with hybrid search, re-ranking, and evaluation.

Updated Oct 31, 2024
Search Tags:#llm#rag#evaluation#retrieval#hybrid-search#prompt-security
One-click install
npx skills add https://github.com/thesammykins/dotfiles --skill llm-rag-engineer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: llm-rag-engineer
Source: https://github.com/thesammykins/dotfiles/tree/main/.agents/skills/llm-rag-engineer
Command: npx skills add https://github.com/thesammykins/dotfiles --skill llm-rag-engineer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Retrieval-Augmented Generation pipelines often struggle with hallucinations, misalignment between retrieved content and answers, and a lack of robust evaluation.

Core Features & Use Cases

  • Design RAG pipelines (Ingestion -> Chunking -> Embedding -> Retrieval -> Generation) to orchestrate data flow.
  • Implement advanced retrieval techniques (Hybrid Search, HyDE expansion, and cross-encoder re-ranking) to boost relevance and faithfulness.
  • Set up evaluation frameworks (Ragas, Arize Phoenix) to quantify hallucination, context fidelity, and uncertainty.
  • Enforce prompt-security best practices to mitigate prompt injection and data leakage.

Quick Start

Configure an end-to-end RAG workflow with evaluation, HyDE expansion, and prompt-security guardrails for a given document store.

Frequently Asked Questions about llm-rag-engineer

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

FAQPage Schema
How do I build a RAG pipeline that reduces hallucinations in answer generation?▼

Build a RAG pipeline with orchestration from ingestion to generation, implementing cross-encoder re-ranking and evaluation frameworks to quantify hallucination, context fidelity, and uncertainty.

How does hybrid search and HyDE expansion improve retrieval augmented generation?▼

Hybrid search and HyDE expansion improve RAG retrieval by combining keyword and semantic search, and generating hypothetical answers to fetch highly relevant document chunks for accurate generation.

What's the best way to evaluate retrieval augmented generation systems for context fidelity?▼

Evaluate RAG systems using frameworks like Ragas and Arize Phoenix to quantify context fidelity, measure hallucination rates, and track uncertainty across retrieval and generation stages.

Can I implement prompt security guardrails to prevent prompt injection in RAG chatbots?▼

Implement prompt-security best practices to enforce guardrails that mitigate prompt injection risks and prevent data leakage within RAG chatbots and AI assistants.

Does this RAG evaluation pipeline support semantic search and AI assistant applications?▼

The RAG pipeline supports semantic search, AI assistants, and chatbots by orchestrating chunking, embedding, and retrieval workflows with integrated evaluation metrics and prompt safety.

Why does my retrieval augmented generation output misalign with retrieved document content?▼

Misalignment between retrieved content and generated answers occurs without cross-encoder re-ranking and robust evaluation metrics to ensure context fidelity and faithful generation.

Related Skills

Explore More Data & Analytics Skills→
NelkitNelkit
1

chromadb-embed

Embed text data into ChromaDB with deterministic IDs and chunking.

Community
Intermediate
48Nauts-Operator48Nauts-Operator
3

rag-implementation

Build Retrieval-Augmented Generation systems with vector databases and semantic search.

Community
Advanced
DaddyElonMusk69DaddyElonMusk69

chroma

Store and retrieve embeddings for semantic search and document retrieval.

Community
Basic
svssdevasvssdeva
1

rag-implementation

Build Retrieval-Augmented Generation systems with semantic search and vector storage.

Community
Advanced