AI AGENTS

haystack

by deepset-ai

agentlead-sourcegithub-agent-framework-repositoriessource:github.comagent-frameworkllm-applicationsragsemantic-searchopen-sourcegithub

Visit Website GitHub

Overview

Open-source AI orchestration framework for building context-engineered, production-ready LLM applications.

Details

Haystack is described by its GitHub source as an open-source AI orchestration framework for building context-engineered, production-ready LLM applications. It supports modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation. The documentation describes Haystack agents as AI agents that retrieve information, generate responses, and perform actions using Haystack components, including a loop-based Agent component that works with chat generators and external tools to solve complex queries iteratively.

When to Use

Build LLM applications that need explicit orchestration across retrieval, routing, memory, and generation. Create AI agents that retrieve information, generate responses, and perform actions with tools. Develop RAG, semantic search, multimodal, or conversational systems using modular pipelines.

Getting Started

Review the GitHub repository at https://github.com/deepset-ai/haystack.
Read the Haystack Agents documentation at https://docs.haystack.deepset.ai/docs/agents.
Inspect the Agent component documentation at https://docs.haystack.deepset.ai/docs/agent.
Build a small pipeline or agent workflow before relying on it in production.

Key Features

•Modular pipelines for LLM application orchestration.
•Agent workflows with control over retrieval, routing, memory, and generation.
•Tool-using, loop-based Agent component for iterative problem solving.
•Supports scalable agents, RAG, multimodal applications, semantic search, and conversational systems.

Capabilities

•ai-orchestration
•agents
•rag
•semantic-search
•conversational-ai
•multimodal-apps

Last updated May 31, 2026