diff --git a/artifact.1/tests/test_core.py b/artifact.1/tests/test_core.py
new file mode 100644
index 0000000..d598450
--- /dev/null
+++ b/artifact.1/tests/test_core.py
@@ -0,0 +1,78 @@
+import pytest
+import pandas as pd
+from scipy.stats import mannwhitneyu
+
+from src.artifact_1 import core
+
+@pytest.fixture
+def sample_run_data_valid():
+    return {
+        "timestamp": "2024-06-10T10:00:00Z",
+        "pinned_flag": True,
+        "runtime": 120.5,
+        "seqcount_retry_count": 2,
+        "mischfenster_dauer": 3.4
+    }
+
+@pytest.fixture
+def sample_run_data_unpinned():
+    return {
+        "timestamp": "2024-06-10T10:02:00Z",
+        "pinned_flag": False,
+        "runtime": 122.0,
+        "seqcount_retry_count": 5,
+        "mischfenster_dauer": 7.2
+    }
+
+def test_calculate_metrics_basic(sample_run_data_valid):
+    result = core.calculate_metrics(sample_run_data_valid)
+    assert isinstance(result, dict)
+    assert "retry_free_rate" in result
+    assert "mischfenster_stats" in result
+    assert "correlations" in result
+    # retry free rate sanity
+    assert 0.0 <= result["retry_free_rate"] <= 1.0
+    # mischfenster_stats contains expected keys
+    stats = result["mischfenster_stats"]
+    for key in ("p50", "p95", "max"):
+        assert key in stats
+    # correlations sanity check
+    assert isinstance(result["correlations"], dict)
+
+def test_calculate_metrics_invalid_input():
+    invalid_data = {
+        "timestamp": "2024-06-10T10:00:00Z",
+        # missing required fields
+    }
+    with pytest.raises((KeyError, ValueError, TypeError)):
+        core.calculate_metrics(invalid_data)
+
+def test_run_mann_whitney_test_basic():
+    data1 = [1.2, 1.4, 1.6, 1.8, 2.0]
+    data2 = [2.1, 2.3, 2.5, 2.7, 2.9]
+    p_value = core.run_mann_whitney_test(data1, data2)
+    assert isinstance(p_value, float)
+    assert 0.0 <= p_value <= 1.0
+
+def test_run_mann_whitney_test_identical_data():
+    data = [1.0, 1.1, 1.2, 1.3]
+    p_value = core.run_mann_whitney_test(data, data.copy())
+    assert pytest.approx(p_value, rel=1e-6) == 1.0
+
+@pytest.mark.parametrize(
+    "data1,data2",
+    [
+        ([1, 2, 3], [4, 5, 6]),
+        ([10.0, 10.5, 10.8], [9.7, 9.8, 9.9]),
+    ]
+)
+def test_run_mann_whitney_test_symmetry(data1, data2):
+    p1 = core.run_mann_whitney_test(data1, data2)
+    p2 = core.run_mann_whitney_test(data2, data1)
+    assert pytest.approx(p1, rel=1e-6) == p2
+
+def test_run_mann_whitney_test_invalid_input():
+    with pytest.raises((ValueError, TypeError)):
+        core.run_mann_whitney_test([], [1, 2, 3])
+    with pytest.raises((ValueError, TypeError)):
+        core.run_mann_whitney_test([1, 2, 3], [None, 5])
\ No newline at end of file