skills.rest
yonatangross
yonatangrossCOMMUNITY
Explore all skills by this author →
𝕏@yonyonizGitHub

ai-native-development

Integrate LLMs, RAG pipelines, embeddings, and vector databases into production apps.

3|Updated Dec 27, 2025
Search Tags:#embeddings#rag#agents#observability#ai-native#vector-databases
One-click install
npx skills add https://github.com/yonatangross/create-yg-app --skill ai-native-development-yonatangross
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-native-development
Source: https://github.com/yonatangross/create-yg-app/tree/main/.claude/skills/ai-native-development
Command: npx skills add https://github.com/yonatangross/create-yg-app --skill ai-native-development-yonatangross

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill solves the challenge of building AI-native applications by integrating LLMs, RAG pipelines, embeddings, vector databases, and agentic workflows.

Core Features & Use Cases

  • RAG pipelines: Retrieve, ground, and answer using embeddings and vector stores with citations.
  • Embeddings & Vector Databases: Produce semantic search, document ingestion, and scalable retrieval across Pinecone, Chroma, Weaviate, or Qdrant.
  • Agentic Workflows: Build ReAct, Tree of Thoughts, and autonomous agents that reason, plan, and act with tools.
  • Local Inference & Observability: Local Ollama-based inference for cost and privacy plus Langfuse-based monitoring.
  • Production Best Practices: Cost optimization, streaming responses, structured outputs, and security considerations.

Quick Start

  • Start a new project using the skill’s patterns, configure Ollama for local inference, wire up a RAG pipeline, enable semantic caching, and integrate your favorite LLMs and tooling.
  • Run a simple end-to-end example: ingest documents, build embeddings, perform a RAG query, and produce a grounded answer with citations.

Frequently Asked Questions about ai-native-development

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

FAQPage Schema
How do I build a RAG pipeline with vector databases for grounded answers?▼

To build a RAG pipeline, ingest documents, generate embeddings, and store them in vector databases like Pinecone, Chroma, Weaviate, or Qdrant. You then retrieve relevant context to ground LLM responses and produce answers with citations.

How do I set up local LLM inference with Ollama for AI-native apps?▼

You can set up local LLM inference with Ollama to ensure privacy and cost control. Configure your application to route requests to Ollama, enabling local inference without relying on external API providers for your AI-native software.

Can I use autonomous agents with tool use in my LLM application?▼

Yes, you can implement agentic workflows using ReAct, Tree of Thoughts, and autonomous agent patterns. These agents reason, plan, and act by utilizing external tools to execute complex tasks within your AI-native application.

What is the best way to monitor and optimize costs for LLM applications?▼

The best way to monitor and optimize LLM application costs is by integrating Langfuse for observability and applying production best practices like semantic caching, streaming responses, and cost-aware deployment patterns.

Do I need structured outputs and streaming for production-grade AI applications?▼

Yes, structured outputs and streaming responses are production best practices required for modular templates and reliable AI-native software. They ensure predictable data formats and real-time user feedback during generation.

How does semantic search work with embeddings and vector stores?▼

Semantic search works by converting documents into embeddings and storing them in vector databases. When a query is processed, the system retrieves documents with similar vector representations, enabling scalable retrieval of contextually relevant information.

Related Skills

Explore More Software Engineering Skills→
constructive-ioconstructive-io

pgvector-rag

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

Official
Advanced
aleonsaaleonsa

llm-engineering

Provide LLM application development patterns for prompts, structured outputs, and observability.

Community
Advanced
haxlyshaxlys

similarity-search-patterns

Perform top-k similarity search over vector embeddings across Pinecone, Qdrant, pgvector, and Weaviate.

Community
Advanced
Nuwanda04Nuwanda04

senior-prompt-engineer

Analyze prompts, evaluate RAG systems, and design agentic workflows.

Community
Advanced