skills.rest
imMamdouhaboammar
imMamdouhaboammarCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

agency-rag-pipeline-engineer

Design and implement retrieval-augmented generation pipelines with vector databases and LLM frameworks.

Updated Jul 24, 2026
Search Tags:#llm#rag#evaluation#retrieval#pipeline#vector-database
One-click install
npx skills add https://github.com/imMamdouhaboammar/kaku-chatgpt-harness --skill agency-rag-pipeline-engineer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agency-rag-pipeline-engineer
Source: https://github.com/imMamdouhaboammar/kaku-chatgpt-harness/tree/main/.agents/skills/engineering-rag-pipeline-engineer
Command: npx skills add https://github.com/imMamdouhaboammar/kaku-chatgpt-harness --skill agency-rag-pipeline-engineer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires langchain, pgvector, asyncpg, openai, ragas, datasets, sentence-transformers, langgraph, sqlalchemy.

What problem does it solve?

This skill addresses the common failure of RAG systems that retrieve irrelevant context or provide unfaithful answers by applying rigorous engineering, evaluation, and architectural best practices.

Core Features & Use Cases

  • Retrieval Architecture: Implements advanced chunking strategies, hybrid search (dense + sparse), and HNSW indexing for optimal precision and latency.
  • Eval-Driven Iteration: Provides a framework for measuring retrieval quality using RAGAS and LangSmith to ensure architectural changes are data-backed.
  • Agentic RAG: Supports complex multi-step retrieval flows, query decomposition, and re-ranking to handle sophisticated user queries.

Quick Start

Use the agency-rag-pipeline-engineer skill to audit my current document chunking strategy and suggest improvements based on my retrieval precision metrics.

Frequently Asked Questions about agency-rag-pipeline-engineer

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

FAQPage Schema
How do I build a production-grade RAG pipeline with async ingestion and metadata filtering?▼

Build a production-grade RAG pipeline by combining asyncpg and SQLAlchemy for high-concurrency async ingestion, using pgvector for metadata-filtered retrieval, and leveraging LangChain to orchestrate the document chunking and embedding workflows.

Why does my retrieval-augmented generation system retrieve irrelevant context or provide unfaithful answers?▼

Irrelevant context and unfaithful answers in retrieval-augmented generation often stem from poorly optimized chunking strategies or lack of hybrid search; implementing systematic evaluation with RAGAS and re-ranking can resolve these architectural failures.

What's the best way to evaluate retrieval quality and RAG performance metrics?▼

Evaluate retrieval quality and RAG performance by integrating RAGAS and LangSmith frameworks to measure retrieval precision and faithfulness, ensuring that architectural changes to your pipeline are fully data-backed.

Can I use pgvector with LangChain for hybrid search and HNSW indexing in my RAG system?▼

Yes, you can use pgvector with LangChain to implement hybrid search combining dense and sparse retrieval, while utilizing HNSW indexing within your vector database to achieve optimal precision and low latency.

How do I implement agentic RAG flows with query decomposition using LangGraph?▼

Implement agentic RAG flows with LangGraph to handle complex multi-step retrieval processes, executing query decomposition and re-ranking to effectively manage sophisticated user queries within your generation pipeline.

Do I need sentence-transformers to optimize embedding strategies for my RAG pipeline?▼

Yes, sentence-transformers are used alongside OpenAI models to systematically evaluate and optimize your embedding strategies, ensuring high retrieval precision before committing documents to the vector database.

Related Skills

Explore More Software Engineering Skills→
patricio0312revpatricio0312rev
5

vector-db-setup

Configure Pinecone, Chroma, pgvector, and Qdrant vector databases for semantic search.

Community
Advanced
ContextJet-aiContextJet-ai
29

monitor-rag-quality

Evaluate RAG pipelines using context precision, recall, faithfulness, and answer relevance.

Official
Intermediate
Miosa-osaMiosa-osa
226

eval-rag

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

Official
Advanced
samelhousseinisamelhousseini

postgres-ai

Automate vector search, RAG, and memory in Azure PostgreSQL with pgvector and azure_ai.

Community
Advanced