diff --git a/integrations/deepeval/pyproject.toml b/integrations/deepeval/pyproject.toml
index 58a9a5e470..01e3011cdc 100644
--- a/integrations/deepeval/pyproject.toml
+++ b/integrations/deepeval/pyproject.toml
@@ -82,6 +82,7 @@ line-length = 120
 [tool.ruff.lint]
 select = [
   "A",
+  "ANN",
   "ARG",
   "B",
   "C",
@@ -110,6 +111,7 @@ select = [
 ignore = [
   # Allow non-abstract empty methods in abstract base classes
   "B027",
+  "ANN401",  # Allow Any - used legitimately for dynamic types and SDK boundaries
   # Allow boolean positional values in function calls, like `dict.get(... True)`
   "FBT003",
   # Ignore checks for possible passwords
@@ -139,7 +141,7 @@ ban-relative-imports = "all"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252"]
+"tests/**/*" = ["PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]
diff --git a/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/evaluator.py b/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/evaluator.py
index f73d4872cd..9f25d642f9 100644
--- a/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/evaluator.py
+++ b/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/evaluator.py
@@ -54,7 +54,7 @@ def __init__(
         self,
         metric: str | DeepEvalMetric,
         metric_params: dict[str, Any] | None = None,
-    ):
+    ) -> None:
         """
         Construct a new DeepEval evaluator.
 
@@ -144,7 +144,7 @@ def from_dict(cls, data: dict[str, Any]) -> "DeepEvalEvaluator":
     def _invoke_deepeval(test_cases: list[LLMTestCase], metric: BaseMetric) -> EvaluationResult:
         return evaluate(test_cases=test_cases, metrics=[metric])
 
-    def _init_backend(self):
+    def _init_backend(self) -> None:
         """
         Initialize the DeepEval backend.
         """
diff --git a/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/metrics.py b/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/metrics.py
index 8575521097..a115ab2c16 100644
--- a/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/metrics.py
+++ b/integrations/deepeval/src/haystack_integrations/components/evaluators/deepeval/metrics.py
@@ -49,7 +49,7 @@ class DeepEvalMetric(Enum):
     #: Inputs - `questions: List[str], contexts: List[List[str]], responses: List[str]`
     CONTEXTUAL_RELEVANCE = "contextual_relevance"
 
-    def __str__(self):
+    def __str__(self) -> str:
         return self.value
 
     @classmethod
@@ -87,7 +87,13 @@ class MetricResult:
     score: float | None = None
     explanation: str | None = None
 
-    def to_dict(self):
+    def to_dict(self) -> dict[str, Any]:
+        """
+        Convert the metric result to a dictionary.
+
+        :returns:
+            A dictionary with the metric result fields.
+        """
         return dataclasses.asdict(self)
 
 
@@ -129,6 +135,23 @@ def new(
         *,
         init_parameters: Mapping[str, type] | None = None,
     ) -> "MetricDescriptor":
+        """
+        Create a new metric descriptor, inferring input parameters from the converter signature.
+
+        :param metric:
+            The metric enum value.
+        :param backend:
+            The DeepEval metric class to instantiate.
+        :param input_converter:
+            Callable that converts Haystack inputs to DeepEval test cases.
+        :param output_converter:
+            Callable that converts DeepEval results to `MetricResult` objects.
+            If ``None``, the default output converter is used.
+        :param init_parameters:
+            Optional mapping of parameter names to types accepted by the backend metric's constructor.
+        :returns:
+            A new `MetricDescriptor` instance.
+        """
         input_converter_signature = inspect.signature(input_converter)
         input_parameters = {}
         for name, param in input_converter_signature.parameters.items():
@@ -158,7 +181,7 @@ class InputConverters:
     """
 
     @staticmethod
-    def _validate_input_elements(**kwargs):
+    def _validate_input_elements(**kwargs: Any) -> None:
         for k, collection in kwargs.items():
             if not isinstance(collection, list):
                 msg = (
@@ -177,6 +200,18 @@ def _validate_input_elements(**kwargs):
 
     @staticmethod
     def validate_input_parameters(metric: DeepEvalMetric, expected: dict[str, Any], received: dict[str, Any]) -> None:
+        """
+        Validate that all expected input parameters are present in the received inputs.
+
+        :param metric:
+            The metric being evaluated, used for error messages.
+        :param expected:
+            Dictionary of expected parameter names to their types.
+        :param received:
+            Dictionary of received parameter names to their values.
+        :raises ValueError:
+            If a required parameter is missing from ``received``.
+        """
         for param, _ in expected.items():
             if param not in received:
                 msg = f"DeepEval evaluator expected input parameter '{param}' for metric '{metric}'"
@@ -186,6 +221,18 @@ def validate_input_parameters(metric: DeepEvalMetric, expected: dict[str, Any],
     def question_context_response(
         questions: list[str], contexts: list[list[str]], responses: list[str]
     ) -> Iterable[LLMTestCase]:
+        """
+        Convert question, context, and response inputs to DeepEval test cases.
+
+        :param questions:
+            List of input questions.
+        :param contexts:
+            List of retrieval context lists, one per question.
+        :param responses:
+            List of model responses, one per question.
+        :returns:
+            An iterable of `LLMTestCase` objects.
+        """
         InputConverters._validate_input_elements(questions=questions, contexts=contexts, responses=responses)
         for q, c, r in zip(questions, contexts, responses, strict=True):  # type: ignore
             test_case = LLMTestCase(input=q, actual_output=r, retrieval_context=c)
@@ -195,6 +242,20 @@ def question_context_response(
     def question_context_response_ground_truth(
         questions: list[str], contexts: list[list[str]], responses: list[str], ground_truths: list[str]
     ) -> Iterable[LLMTestCase]:
+        """
+        Convert question, context, response, and ground truth inputs to DeepEval test cases.
+
+        :param questions:
+            List of input questions.
+        :param contexts:
+            List of retrieval context lists, one per question.
+        :param responses:
+            List of model responses, one per question.
+        :param ground_truths:
+            List of expected (ground truth) responses, one per question.
+        :returns:
+            An iterable of `LLMTestCase` objects.
+        """
         InputConverters._validate_input_elements(questions=questions, contexts=contexts, responses=responses)
         for q, c, r, gt in zip(questions, contexts, responses, ground_truths, strict=True):  # type: ignore
             test_case = LLMTestCase(input=q, actual_output=r, retrieval_context=c, expected_output=gt)
@@ -212,6 +273,15 @@ class OutputConverters:
     def default(
         metric: DeepEvalMetric,
     ) -> Callable[[TestResult], list[MetricResult]]:
+        """
+        Return the default output converter for a given metric.
+
+        :param metric:
+            The metric for which to create the converter.
+        :returns:
+            A callable that converts a `TestResult` to a list of `MetricResult` objects.
+        """
+
         def inner(output: TestResult, metric: DeepEvalMetric) -> list[MetricResult]:
             metric_name = str(metric)
             assert output.metrics_data