Dynamiq Docs
  • Welcome to Dynamiq
  • Low-Code Builder
    • Chat
    • Basics
    • Connecting Nodes
    • Conditional Nodes and Multiple Outputs
    • Input and Output Transformers
    • Error Handling and Retries
    • LLM Nodes
    • Validator Nodes
    • RAG Nodes
      • Indexing Workflow
        • Pre-processing Nodes
        • Document Splitting
        • Document Embedders
        • Document Writers
      • Inference RAG workflow
        • Text embedders
        • Document retrievers
          • Complex retrievers
        • LLM Answer Generators
    • LLM Agents
      • Basics
      • Guide to Implementing LLM Agents: ReAct and Simple Agents
      • Guide to Agent Orchestration: Linear and Adaptive Orchestrators
      • Guide to Advanced Agent Orchestration: Graph Orchestrator
    • Audio and voice
    • Tools and External Integrations
    • Python Code in Workflows
    • Memory
    • Guardrails
  • Deployments
    • Workflows
      • Tracing Workflow Execution
    • LLMs
      • Fine-tuned Adapters
      • Supported Models
    • Vector Databases
  • Prompts
    • Prompt Playground
  • Connections
  • LLM Fine-tuning
    • Basics
    • Using Adapters
    • Preparing Data
    • Supported Models
    • Parameters Guide
  • Knowledge Bases
  • Evaluations
    • Metrics
      • LLM-as-a-Judge
      • Predefined metrics
        • Faithfulness
        • Context Precision
        • Context Recall
        • Factual Correctness
        • Answer Correctness
      • Python Code Metrics
    • Datasets
    • Evaluation Runs
    • Examples
      • Build Accurate vs. Inaccurate Workflows
  • Examples
    • Building a Search Assistant
      • Approach 1: Single Agent with a Defined Role
      • Approach 2: Adaptive Orchestrator with Multiple Agents
      • Approach 3: Custom Logic Pipeline with a Straightforward Workflow
    • Building a Code Assistant
  • Platform Settings
    • Access Keys
    • Organizations
    • Settings
    • Billing
  • On-premise Deployment
    • AWS
    • IBM
  • Support Center
Powered by GitBook
On this page
  • Overview
  • Understanding Nodes
  • Common Traits of Nodes
  • Connecting Nodes
  • Categories of Nodes
  1. Low-Code Builder

Basics

Overview

The Low-Code Builder in Dynamiq is a powerful and intuitive interface designed to help users create and manage AI workflows with minimal coding effort. It provides a drag-and-drop environment where you can build complex workflows by connecting various nodes, each representing a specific action or operation. The builder supports a wide range of pre-built nodes that can be customized and linked together to automate tasks, integrate with external services, and innovate with AI-driven applications.

Understanding Nodes

Nodes are the fundamental building blocks of workflows in the Low-Code Builder. Each node represents a discrete action or operation, e.g. processing data, making an API call, or interacting with an LLM. Nodes can be connected to form a workflow, where the output of one node serves as the input for the next.

Common Traits of Nodes

  • Configuration Tab: Each node can be configured to meet the specific needs of your workflow. The Configuration tab allows you to set up basic parameters and customize the node's behavior.

  • Input Tab: The Input tab lets you define how the node receives input data. You can filter and transform the input using JSONPath syntax, which allows you to work with specific parts of the input data.

    • The JSONPath syntax follows this format: $.{node_name}.output.{parameter_name}, which ensures that you can access specific parameters from the output of previous nodes.

  • Output Tab: The Output tab controls how the node's output is processed before being passed to the next node in the workflow. Like the Input tab, you can filter and transform the output using JSONPath.

  • Error Handling Tab: The Error Handling tab provides options to manage how the node responds to errors. You can configure retry intervals, set a maximum number of retry attempts, define a backoff rate for retries, and specify a timeout duration for the node.

  • Deleting Nodes: To delete a node, select the node you wish to delete and then press the backspace button on your keyboard.

Connecting Nodes

Nodes can be connected by linking their input and output circles. Each node typically has a primary connection point at the top level, but some nodes have additional, smaller connection points for more specific inputs. When these secondary connections are used, they automatically add input transformers, enabling complex data manipulations as part of the workflow.

The input/output transformers in Dynamiq work similarly to those in AWS Step Functions, allowing you to transform data as it moves between nodes.

Categories of Nodes

Nodes in the Low-Code Builder are categorized based on their function. Below are the primary categories and some examples:

  • Rankers: LLM Document Ranker, Time Weighted Document Ranker

  • Vectorization: OpenAI Document Embedder, OpenAI Text Embedder

  • Storages: Weaviate Retriever, Pinecone Retriever, Chroma Retriever

  • Audio: Whisper, ElevenLabs STS, ElevenLabs TTS

  • Validators: Regex Match, Valid Choices, Valid JSON, Valid Python

  • Pre-Processing: Unstructured Converter

  • Chunking: Document Splitter

  • LLM Models: OpenAI, Anthropic, Cohere, Gemini, AWS Bedrock, Mistral, and more

  • Tools: Tavily, ZenRows, FireCrawl, ScaleSerp, Python, HTTP API Call

  • Logic Operators: Choice

  • Orchestrators: Linear Orchestrator, Adaptive Orchestrator

  • Agents: ReAct Agent, Simple Agent

Each category serves a different purpose, allowing you to build workflows tailored to your specific needs.

PreviousChatNextConnecting Nodes

Last updated 2 months ago