diff --git a/.gitignore b/.gitignore
index 1156976f..48bed481 100644
--- a/.gitignore
+++ b/.gitignore
@@ -11,7 +11,7 @@ wheels/
 
 # tests
 .coverage*
-evaluations/data/
+src/evaluations/data/
 tests/data/
 .pytest_cache/
 .ruff_cache/
diff --git a/docs/benchmarks.md b/docs/benchmarks.md
index 07f0c6df..33921382 100644
--- a/docs/benchmarks.md
+++ b/docs/benchmarks.md
@@ -3,7 +3,7 @@
 We use the [repliqa](https://huggingface.co/datasets/ServiceNow/repliqa) dataset for the evaluation of `haiku.rag`.
 
 You can perform your own evaluations with the Typer CLI in
-`evaluations/benchmark.py`, for example `python -m evaluations.benchmark repliqa`.
+`src/evaluations/benchmark.py`, for example `cd src && python -m evaluations.benchmark repliqa`.
 The evaluation flow is orchestrated with
 [`pydantic-evals`](https://github.com/pydantic/pydantic-ai/tree/main/libs/pydantic-evals),
 which we leverage for dataset management, scoring, and report generation.
diff --git a/pyproject.toml b/pyproject.toml
index 09727f05..9c9e9c83 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -49,7 +49,7 @@ requires = ["hatchling"]
 build-backend = "hatchling.build"
 
 [tool.hatch.build]
-exclude = ["/docs", "/evaluations", "/tests", "/.github"]
+exclude = ["/docs", "/tests", "/.github"]
 
 [tool.hatch.build.targets.wheel]
 packages = ["src/haiku"]
diff --git a/evaluations/__init__.py b/src/evaluations/__init__.py
similarity index 100%
rename from evaluations/__init__.py
rename to src/evaluations/__init__.py
diff --git a/evaluations/benchmark.py b/src/evaluations/benchmark.py
similarity index 100%
rename from evaluations/benchmark.py
rename to src/evaluations/benchmark.py
diff --git a/evaluations/config.py b/src/evaluations/config.py
similarity index 100%
rename from evaluations/config.py
rename to src/evaluations/config.py
diff --git a/evaluations/datasets/__init__.py b/src/evaluations/datasets/__init__.py
similarity index 100%
rename from evaluations/datasets/__init__.py
rename to src/evaluations/datasets/__init__.py
diff --git a/evaluations/datasets/repliqa.py b/src/evaluations/datasets/repliqa.py
similarity index 100%
rename from evaluations/datasets/repliqa.py
rename to src/evaluations/datasets/repliqa.py
diff --git a/evaluations/datasets/wix.py b/src/evaluations/datasets/wix.py
similarity index 100%
rename from evaluations/datasets/wix.py
rename to src/evaluations/datasets/wix.py
diff --git a/evaluations/llm_judge.py b/src/evaluations/llm_judge.py
similarity index 100%
rename from evaluations/llm_judge.py
rename to src/evaluations/llm_judge.py