Circle V2 API Docs
    Preparing search index...

    Type Alias RunChartReviewEvalExperimentOptions

    type RunChartReviewEvalExperimentOptions = {
        datasetName: string;
        experimentPrefix?: string;
        extraEvaluators?: ChartReviewEvaluator[];
        maxConcurrency?: number;
        metadata?: Record<string, unknown>;
        numRepetitions?: number;
        patientId?: PatientId;
        settings?: Partial<LegacyChartReviewSettings>;
        splits?: string[];
    }
    Index

    Properties

    datasetName: string

    LangSmith dataset name, e.g. "Chart Review Run Evals".

    experimentPrefix?: string

    Prefix for the LangSmith experiment name.

    extraEvaluators?: ChartReviewEvaluator[]

    Extra scorers appended to the built-in status_match + no_fabricated_documents — the caller threads in the runner-agnostic evaluators (answer_correctness, true_pass) so a V2 experiment is scored by the same metrics as V3. Kept generic (no @repo/ai-agents dep here); the CLI wires them together.

    maxConcurrency?: number

    Concurrency for target runs (each is an OpenAI-bound chart-review question).

    metadata?: Record<string, unknown>

    Extra metadata stamped on the experiment (git commit, prompt tag, ...) for comparability.

    numRepetitions?: number

    Run every example N times to measure verdict variance — temp-0 is not fully deterministic.

    patientId?: PatientId

    Sugar for splits: ["p<patientId>"] — this cloned patient's examples.

    settings?: Partial<LegacyChartReviewSettings>

    Baseline settings overrides — part of the "application version" being evaluated.

    splits?: string[]

    Dataset splits to evaluate, LITERALLY — ["p<patientId>"] runs that patient's examples regardless of review state (so a freshly-added incoming run can be evaluated before promotion). Multiple splits are a UNION. Omit (and omit patientId) to run the golden split — the baseline membership.