Package software.amazon.awscdk.services.bedrockagentcore

Class LlmAsAJudgeOptions.Builder

java.lang.Object
software.amazon.awscdk.services.bedrockagentcore.LlmAsAJudgeOptions.Builder
All Implemented Interfaces:
software.amazon.jsii.Builder<LlmAsAJudgeOptions>
Enclosing interface:
LlmAsAJudgeOptions
@Stability(Stable) public static final class LlmAsAJudgeOptions.Builder extends Object implements software.amazon.jsii.Builder<LlmAsAJudgeOptions>
A builder for LlmAsAJudgeOptions

  • Constructor Details

    • Builder

      public Builder()

  • Method Details

    • instructions

      @Stability(Stable) public LlmAsAJudgeOptions.Builder instructions(String instructions)
      Sets the value of LlmAsAJudgeOptions.getInstructions()
      Parameters:
      instructions - The evaluation instructions that guide the language model in assessing agent performance. This parameter is required. These instructions define the evaluation criteria, context, and expected behavior. Instructions must contain placeholders appropriate for the evaluation level (e.g., {context}, {available_tools} for SESSION level).

      Note: Evaluators using reference-input placeholders (e.g., {expected_tool_trajectory}, {assertions}, {expected_response}) are only compatible with on-demand evaluation, not online evaluation.

      Returns:
      this

    • modelId

      @Stability(Stable) public LlmAsAJudgeOptions.Builder modelId(String modelId)
      Sets the value of LlmAsAJudgeOptions.getModelId()
      Parameters:
      modelId - The identifier of the Amazon Bedrock model to use for evaluation. This parameter is required. Accepts standard model IDs (e.g., 'anthropic.claude-sonnet-4-6') and cross-region inference profile IDs with region prefixes (e.g., 'us.anthropic.claude-sonnet-4-6', 'eu.anthropic.claude-sonnet-4-6').
      Returns:
      this

    • ratingScale

      @Stability(Stable) public LlmAsAJudgeOptions.Builder ratingScale(EvaluatorRatingScale ratingScale)
      Sets the value of LlmAsAJudgeOptions.getRatingScale()
      Parameters:
      ratingScale - The rating scale that defines how the evaluator should score agent performance. This parameter is required.
      Returns:
      this

    • additionalModelRequestFields

      @Stability(Stable) public LlmAsAJudgeOptions.Builder additionalModelRequestFields(Map<String,? extends Object> additionalModelRequestFields)
      Sets the value of LlmAsAJudgeOptions.getAdditionalModelRequestFields()
      Parameters:
      additionalModelRequestFields - Additional model-specific request fields.
      Returns:
      this

    • inferenceConfig

      @Stability(Stable) public LlmAsAJudgeOptions.Builder inferenceConfig(EvaluatorInferenceConfig inferenceConfig)
      Sets the value of LlmAsAJudgeOptions.getInferenceConfig()
      Parameters:
      inferenceConfig - Optional inference configuration parameters that control model behavior during evaluation. When not specified, the foundation model uses its own default values for maxTokens, temperature, and topP.
      Returns:
      this

    • build

      @Stability(Stable) public LlmAsAJudgeOptions build()
      Builds the configured instance.
      Specified by:
      build in interface software.amazon.jsii.Builder<LlmAsAJudgeOptions>
      Returns:
      a new instance of LlmAsAJudgeOptions
      Throws:
      NullPointerException - if any required attribute was not provided