deepset-ai/haystack

Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. Design modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation. Built for scalable agents, RAG, multimodal applications, semantic search, and conversational systems.

What it solves

Haystack is an AI orchestration framework designed to move LLM applications from prototype to production. It provides the tools to build scalable Retrieval-Augmented Generation (RAG) systems, autonomous agents, and semantic search applications while maintaining explicit control over how data is handled and processed.

How it works

The framework uses a modular architecture based on pipelines and agent workflows. Users can combine built-in components for indexing, retrieval, routing, and generation, or create custom components. It is model- and vendor-agnostic, allowing developers to swap LLMs (such as OpenAI, Anthropic, or local models) and infrastructure components without rewriting the entire system. It also supports native async operations and token streaming.

Who it’s for

It is built for Python developers and AI engineers who need to build production-ready AI systems that require deep customization, traceability, and the ability to scale across different model providers.

Highlights

  • Production-Ready Agents: Includes lifecycle hooks for guardrails, monitoring for token usage and cost, and a SkillToolset for progressive skill discovery.
  • Context Engineering: Offers explicit control over how information is retrieved, ranked, filtered, and routed before reaching the model.
  • Vendor Agnostic: Integrates with a wide range of providers including Hugging Face, Google, Azure OpenAI, and AWS Bedrock.
  • Extensible Ecosystem: Allows the creation and sharing of custom components through a consistent interface.

Related

  • Project
    dynamiq-ai/dynamiq

    Dynamiq is an all-in-one Python framework for building AI applications that combine LLMs, agents, and RAG workflows. It provides orchestration components to easily define, run, and manage complex AI pipelines.

  • Project
    agentscope-ai/agentscope

    A production-ready agent framework that provides an SDK and a FastAPI backend to build multi-tenant, multi-session agentic applications leveraging LLM reasoning and tool use.

  • Project
    EpicStaff/EpicStaff

    EpicStaff is a self‑hosted, source‑available platform that lets teams design, run, and audit multi‑agent AI workflows via a visual node editor built on Django. It supports custom Python tools, LLM/RAG integration, persistent memory, webhook triggers, and enterprise security, giving operations teams ownership of AI processes while engineers handle system integration.

  • Project
    garrytan/gstack

    gstack is an open-source framework that turns AI agents into a structured engineering team by providing specialized slash commands for product planning, security auditing, QA, and release management.

  • Project
    synapseorch-ai/synapse-ai

    A multi-agent orchestration platform that allows users to build deterministic AI workflows by wiring together agents, tools, and various LLMs into scalable pipelines.