diff --git a/extensions/positron-python/pythonFiles/positron/positron_ipykernel/data_explorer.py b/extensions/positron-python/pythonFiles/positron/positron_ipykernel/data_explorer.py
index 7fe12776473..05eccba168b 100644
--- a/extensions/positron-python/pythonFiles/positron/positron_ipykernel/data_explorer.py
+++ b/extensions/positron-python/pythonFiles/positron/positron_ipykernel/data_explorer.py
@@ -31,7 +31,7 @@
     ColumnProfileRequestType,
     ColumnProfileResult,
     ColumnSchema,
-    ColumnSchemaTypeDisplay,
+    ColumnDisplayType,
     ColumnSortKey,
     DataExplorerBackendMessageContent,
     DataExplorerFrontendEvent,
@@ -48,6 +48,9 @@
     SearchSchemaResult,
     SetRowFiltersRequest,
     SetSortColumnsRequest,
+    SummaryStatsBoolean,
+    SummaryStatsNumber,
+    SummaryStatsString,
     TableData,
     TableSchema,
     TableShape,
@@ -294,6 +297,14 @@ def __init__(
         # performance.
         self._search_schema_last_result: Optional[Tuple[str, List[ColumnSchema]]] = None
 
+        # Putting this here rather than in the class body before
+        # Python < 3.10 has fussier rules about staticmethods
+        self._SUMMARIZERS = {
+            ColumnDisplayType.Boolean: self._summarize_boolean,
+            ColumnDisplayType.Number: self._summarize_number,
+            ColumnDisplayType.String: self._summarize_string,
+        }
+
     def invalidate_computations(self):
         self.filtered_indices = self.view_indices = None
         self._need_recompute = True
@@ -337,10 +348,7 @@ def _get_schema(self, column_start: int, num_columns: int) -> TableSchema:
             column_start,
             min(column_start + num_columns, len(self.table.columns)),
         ):
-            column_raw_name = self.table.columns[column_index]
-            column_name = str(column_raw_name)
-
-            col_schema = self._get_single_column_schema(column_index, column_name)
+            col_schema = self._get_single_column_schema(column_index)
             column_schemas.append(col_schema)
 
         return TableSchema(columns=column_schemas)
@@ -376,7 +384,7 @@ def _search_schema_get_matches(self, search_term: str) -> List[ColumnSchema]:
             if search_term not in column_name.lower():
                 continue
 
-            col_schema = self._get_single_column_schema(column_index, column_name)
+            col_schema = self._get_single_column_schema(column_index)
             matches.append(col_schema)
 
         return matches
@@ -388,7 +396,10 @@ def _get_inferred_dtype(self, column_index: int):
             self._inferred_dtypes[column_index] = infer_dtype(self.table.iloc[:, column_index])
         return self._inferred_dtypes[column_index]
 
-    def _get_single_column_schema(self, column_index: int, column_name: str):
+    def _get_single_column_schema(self, column_index: int):
+        column_raw_name = self.table.columns[column_index]
+        column_name = str(column_raw_name)
+
         # TODO: pandas MultiIndex columns
         # TODO: time zone for datetimetz datetime64[ns] types
         dtype = self.dtypes.iloc[column_index]
@@ -405,7 +416,7 @@ def _get_single_column_schema(self, column_index: int, column_name: str):
             column_name=column_name,
             column_index=column_index,
             type_name=type_name,
-            type_display=ColumnSchemaTypeDisplay(type_display),
+            type_display=ColumnDisplayType(type_display),
         )
 
     def _get_data_values(
@@ -431,7 +442,8 @@ def _get_data_values(
 
         if self.view_indices is not None:
             # If the table is either filtered or sorted, use a slice
-            # the view_indices to select the virtual range of values for the grid
+            # the view_indices to select the virtual range of values
+            # for the grid
             view_slice = self.view_indices[row_start : row_start + num_rows]
             columns = [col.take(view_slice) for col in columns]
             indices = self.table.index.take(view_slice)
@@ -544,9 +556,12 @@ def _eval_filter(self, filt: RowFilter):
                 elif params.type == SearchFilterParamsType.EndsWith:
                     mask = col.str.endswith(term)
 
+        assert mask is not None
+
         # Nulls are possible in the mask, so we just fill them if any
         if mask.dtype != bool:
-            mask = mask.fillna(False)
+            mask[mask.isna()] = False
+            mask = mask.astype(bool)
 
         return mask.to_numpy()
 
@@ -602,7 +617,57 @@ def _prof_null_count(self, column_index: int):
         return self._get_column(column_index).isnull().sum()
 
     def _prof_summary_stats(self, column_index: int):
-        raise NotImplementedError
+        col_schema = self._get_single_column_schema(column_index)
+        col = self._get_column(column_index)
+
+        ui_type = col_schema.type_display
+        handler = self._SUMMARIZERS.get(ui_type)
+
+        if handler is None:
+            # Return nothing for types we don't yet know how to summarize
+            return ColumnSummaryStats(type_display=ui_type)
+        else:
+            return handler(col)
+
+    @staticmethod
+    def _summarize_number(col: "pd.Series"):
+        min_value = col.min()
+        max_value = col.max()
+        mean = col.mean()
+        median = col.median()
+        stdev = col.std()
+
+        return ColumnSummaryStats(
+            type_display=ColumnDisplayType.Number,
+            number_stats=SummaryStatsNumber(
+                min_value=str(min_value),
+                max_value=str(max_value),
+                mean=str(mean),
+                median=str(median),
+                stdev=str(stdev),
+            ),
+        )
+
+    @staticmethod
+    def _summarize_string(col: "pd.Series"):
+        num_empty = (col.str.len() == 0).sum()
+        num_unique = col.nunique()
+
+        return ColumnSummaryStats(
+            type_display=ColumnDisplayType.String,
+            string_stats=SummaryStatsString(num_empty=num_empty, num_unique=num_unique),
+        )
+
+    @staticmethod
+    def _summarize_boolean(col: "pd.Series"):
+        null_count = col.isnull().sum()
+        true_count = col.sum()
+        false_count = len(col) - true_count - null_count
+
+        return ColumnSummaryStats(
+            type_display=ColumnDisplayType.Boolean,
+            boolean_stats=SummaryStatsBoolean(true_count=true_count, false_count=false_count),
+        )
 
     def _prof_freq_table(self, column_index: int):
         raise NotImplementedError
diff --git a/extensions/positron-python/pythonFiles/positron/positron_ipykernel/data_explorer_comm.py b/extensions/positron-python/pythonFiles/positron/positron_ipykernel/data_explorer_comm.py
index 5b832565a3f..de002df2025 100644
--- a/extensions/positron-python/pythonFiles/positron/positron_ipykernel/data_explorer_comm.py
+++ b/extensions/positron-python/pythonFiles/positron/positron_ipykernel/data_explorer_comm.py
@@ -18,9 +18,9 @@
 
 
 @enum.unique
-class ColumnSchemaTypeDisplay(str, enum.Enum):
+class ColumnDisplayType(str, enum.Enum):
     """
-    Possible values for TypeDisplay in ColumnSchema
+    Possible values for ColumnDisplayType
     """
 
     Number = "number"
@@ -202,7 +202,7 @@ class ColumnSchema(BaseModel):
         description="Exact name of data type used by underlying table",
     )
 
-    type_display: ColumnSchemaTypeDisplay = Field(
+    type_display: ColumnDisplayType = Field(
         description="Canonical Positron display name of data type",
     )
 
@@ -387,7 +387,33 @@ class ColumnProfileResult(BaseModel):
 
 class ColumnSummaryStats(BaseModel):
     """
-    ColumnSummaryStats in Schemas
+    Profile result containing summary stats for a column based on the data
+    type
+    """
+
+    type_display: ColumnDisplayType = Field(
+        description="Canonical Positron display name of data type",
+    )
+
+    number_stats: Optional[SummaryStatsNumber] = Field(
+        default=None,
+        description="Statistics for a numeric data type",
+    )
+
+    string_stats: Optional[SummaryStatsString] = Field(
+        default=None,
+        description="Statistics for a string-like data type",
+    )
+
+    boolean_stats: Optional[SummaryStatsBoolean] = Field(
+        default=None,
+        description="Statistics for a boolean data type",
+    )
+
+
+class SummaryStatsNumber(BaseModel):
+    """
+    SummaryStatsNumber in Schemas
     """
 
     min_value: str = Field(
@@ -398,24 +424,44 @@ class ColumnSummaryStats(BaseModel):
         description="Maximum value as string",
     )
 
-    mean_value: Optional[str] = Field(
-        default=None,
+    mean: str = Field(
         description="Average value as string",
     )
 
-    median: Optional[str] = Field(
-        default=None,
+    median: str = Field(
         description="Sample median (50% value) value as string",
     )
 
-    q25: Optional[str] = Field(
-        default=None,
-        description="25th percentile value as string",
+    stdev: str = Field(
+        description="Sample standard deviation as a string",
     )
 
-    q75: Optional[str] = Field(
-        default=None,
-        description="75th percentile value as string",
+
+class SummaryStatsBoolean(BaseModel):
+    """
+    SummaryStatsBoolean in Schemas
+    """
+
+    true_count: int = Field(
+        description="The number of non-null true values",
+    )
+
+    false_count: int = Field(
+        description="The number of non-null false values",
+    )
+
+
+class SummaryStatsString(BaseModel):
+    """
+    SummaryStatsString in Schemas
+    """
+
+    num_empty: int = Field(
+        description="The number of empty / length-zero values",
+    )
+
+    num_unique: int = Field(
+        description="The exact number of distinct values",
     )
 
 
@@ -560,7 +606,7 @@ class SearchSchemaParams(BaseModel):
     """
 
     search_term: str = Field(
-        description="Substring to match for (currently case insensitive",
+        description="Substring to match for (currently case insensitive)",
     )
 
     start_index: int = Field(
@@ -798,6 +844,12 @@ class SchemaUpdateParams(BaseModel):
 
 ColumnSummaryStats.update_forward_refs()
 
+SummaryStatsNumber.update_forward_refs()
+
+SummaryStatsBoolean.update_forward_refs()
+
+SummaryStatsString.update_forward_refs()
+
 ColumnHistogram.update_forward_refs()
 
 ColumnFrequencyTable.update_forward_refs()
diff --git a/extensions/positron-python/pythonFiles/positron/positron_ipykernel/tests/test_data_explorer.py b/extensions/positron-python/pythonFiles/positron/positron_ipykernel/tests/test_data_explorer.py
index 64d0bab6e3d..26c04fb157b 100644
--- a/extensions/positron-python/pythonFiles/positron/positron_ipykernel/tests/test_data_explorer.py
+++ b/extensions/positron-python/pythonFiles/positron/positron_ipykernel/tests/test_data_explorer.py
@@ -13,6 +13,7 @@
 from ..access_keys import encode_access_key
 from ..data_explorer import COMPARE_OPS, DataExplorerService
 from ..data_explorer_comm import (
+    ColumnDisplayType,
     RowFilter,
     ColumnProfileResult,
     ColumnSchema,
@@ -426,7 +427,7 @@ def compare_tables(self, table_id: str, expected_id: str, table_shape: tuple):
 
 
 @pytest.fixture()
-def de_fixture(de_service: DataExplorerService):
+def dxf(de_service: DataExplorerService):
     return DataExplorerFixture(de_service)
 
 
@@ -434,8 +435,8 @@ def _wrap_json(model: Type[BaseModel], data: JsonRecords):
     return [model(**d).dict() for d in data]
 
 
-def test_pandas_get_state(de_fixture: DataExplorerFixture):
-    result = de_fixture.get_state("simple")
+def test_pandas_get_state(dxf: DataExplorerFixture):
+    result = dxf.get_state("simple")
     assert result["table_shape"]["num_rows"] == 5
     assert result["table_shape"]["num_columns"] == 6
 
@@ -444,17 +445,15 @@ def test_pandas_get_state(de_fixture: DataExplorerFixture):
         {"column_index": 1, "ascending": False},
     ]
     filters = [_compare_filter(0, ">", 0), _compare_filter(0, "<", 5)]
-    de_fixture.set_sort_columns("simple", sort_keys=sort_keys)
-    de_fixture.set_row_filters("simple", filters=filters)
+    dxf.set_sort_columns("simple", sort_keys=sort_keys)
+    dxf.set_row_filters("simple", filters=filters)
 
-    result = de_fixture.get_state("simple")
+    result = dxf.get_state("simple")
     assert result["sort_keys"] == sort_keys
     assert result["row_filters"] == [RowFilter(**f) for f in filters]
 
 
-def test_pandas_get_schema(de_fixture: DataExplorerFixture):
-    dxf = de_fixture
-
+def test_pandas_get_schema(dxf: DataExplorerFixture):
     result = dxf.get_schema("simple", 0, 100)
 
     full_schema = [
@@ -524,9 +523,7 @@ def test_pandas_get_schema(de_fixture: DataExplorerFixture):
     assert result["columns"] == _wrap_json(ColumnSchema, bigger_schema[10:20])
 
 
-def test_pandas_wide_schemas(de_fixture: DataExplorerFixture):
-    dxf = de_fixture
-
+def test_pandas_wide_schemas(dxf: DataExplorerFixture):
     arr = np.arange(10).astype(object)
 
     ncols = 10000
@@ -550,9 +547,7 @@ def test_pandas_wide_schemas(de_fixture: DataExplorerFixture):
             assert left == right
 
 
-def test_pandas_search_schema(de_fixture: DataExplorerFixture):
-    dxf = de_fixture
-
+def test_pandas_search_schema(dxf: DataExplorerFixture):
     # Make a few thousand column names we can search for
     column_names = [
         f"{prefix}_{i}"
@@ -591,8 +586,8 @@ def _trim_whitespace(columns):
     return [[x.strip() for x in column] for column in columns]
 
 
-def test_pandas_get_data_values(de_fixture: DataExplorerFixture):
-    result = de_fixture.get_data_values(
+def test_pandas_get_data_values(dxf: DataExplorerFixture):
+    result = dxf.get_data_values(
         "simple",
         row_start_index=0,
         num_rows=20,
@@ -621,14 +616,12 @@ def test_pandas_get_data_values(de_fixture: DataExplorerFixture):
     assert result["row_labels"] == [["0", "1", "2", "3", "4"]]
 
     # Edge cases: request beyond end of table
-    response = de_fixture.get_data_values(
-        "simple", row_start_index=5, num_rows=10, column_indices=[0]
-    )
+    response = dxf.get_data_values("simple", row_start_index=5, num_rows=10, column_indices=[0])
     assert response["columns"] == [[]]
 
     # Issue #2149 -- return empty result when requesting non-existent
     # column indices
-    response = de_fixture.get_data_values(
+    response = dxf.get_data_values(
         "simple", row_start_index=0, num_rows=5, column_indices=[2, 3, 4, 5]
     )
     assert _trim_whitespace(response["columns"]) == expected_columns[2:]
@@ -638,7 +631,7 @@ def test_pandas_get_data_values(de_fixture: DataExplorerFixture):
     # to request non-existent column indices, disable this test
 
     # with pytest.raises(IndexError):
-    #     de_fixture.get_data_values(
+    #     dxf.get_data_values(
     #         "simple", row_start_index=0, num_rows=10, column_indices=[4]
     #     )
 
@@ -690,8 +683,7 @@ def _set_member_filter(column_index, values, inclusive=True):
     )
 
 
-def test_pandas_filter_between(de_fixture: DataExplorerFixture):
-    dxf = de_fixture
+def test_pandas_filter_between(dxf: DataExplorerFixture):
     df = SIMPLE_PANDAS_DF
     column = "a"
     column_index = df.columns.get_loc(column)
@@ -719,7 +711,7 @@ def test_pandas_filter_between(de_fixture: DataExplorerFixture):
         )
 
 
-def test_pandas_filter_compare(de_fixture: DataExplorerFixture):
+def test_pandas_filter_compare(dxf: DataExplorerFixture):
     # Just use the 'a' column to smoke test comparison filters on
     # integers
     table_name = "simple"
@@ -731,23 +723,23 @@ def test_pandas_filter_compare(de_fixture: DataExplorerFixture):
     for op, op_func in COMPARE_OPS.items():
         filt = _compare_filter(column_index, op, str(compare_value))
         expected_df = df[op_func(df[column], compare_value)]
-        de_fixture.check_filter_case(df, [filt], expected_df)
+        dxf.check_filter_case(df, [filt], expected_df)
 
     # TODO(wesm): move these tests to their own test case
 
     # Test that passing empty filter set resets to unfiltered state
     filt = _compare_filter(column_index, "<", str(compare_value))
-    _ = de_fixture.set_row_filters(table_name, filters=[filt])
-    response = de_fixture.set_row_filters(table_name, filters=[])
+    _ = dxf.set_row_filters(table_name, filters=[filt])
+    response = dxf.set_row_filters(table_name, filters=[])
     assert response == FilterResult(selected_num_rows=len(df))
 
     # register the whole table to make sure the filters are really cleared
     ex_id = guid()
-    de_fixture.register_table(ex_id, df)
-    de_fixture.compare_tables(table_name, ex_id, df.shape)
+    dxf.register_table(ex_id, df)
+    dxf.compare_tables(table_name, ex_id, df.shape)
 
 
-def test_pandas_filter_is_null_not_null(de_fixture: DataExplorerFixture):
+def test_pandas_filter_is_null_not_null(dxf: DataExplorerFixture):
     df = SIMPLE_PANDAS_DF
     b_is_null = _filter("is_null", 1)
     b_not_null = _filter("not_null", 1)
@@ -760,10 +752,10 @@ def test_pandas_filter_is_null_not_null(de_fixture: DataExplorerFixture):
     ]
 
     for filter_set, expected_df in cases:
-        de_fixture.check_filter_case(df, filter_set, expected_df)
+        dxf.check_filter_case(df, filter_set, expected_df)
 
 
-def test_pandas_filter_set_membership(de_fixture: DataExplorerFixture):
+def test_pandas_filter_set_membership(dxf: DataExplorerFixture):
     df = SIMPLE_PANDAS_DF
 
     cases = [
@@ -778,11 +770,10 @@ def test_pandas_filter_set_membership(de_fixture: DataExplorerFixture):
     ]
 
     for filter_set, expected_df in cases:
-        de_fixture.check_filter_case(df, filter_set, expected_df)
+        dxf.check_filter_case(df, filter_set, expected_df)
 
 
-def test_pandas_filter_search(de_fixture: DataExplorerFixture):
-    dxf = de_fixture
+def test_pandas_filter_search(dxf: DataExplorerFixture):
     df = pd.DataFrame(
         {
             "a": ["foo1", "foo2", None, "2FOO", "FOO3", "bar1", "2BAR"],
@@ -841,7 +832,8 @@ def test_pandas_filter_search(de_fixture: DataExplorerFixture):
     ]
 
     for search_type, column_index, term, cs, mask in cases:
-        ex_table = df[mask.fillna(False)]
+        mask[mask.isna()] = False
+        ex_table = df[mask.astype(bool)]
         dxf.check_filter_case(
             df,
             [
@@ -856,7 +848,7 @@ def test_pandas_filter_search(de_fixture: DataExplorerFixture):
         )
 
 
-def test_pandas_set_sort_columns(de_fixture: DataExplorerFixture):
+def test_pandas_set_sort_columns(dxf: DataExplorerFixture):
     tables = {
         "df1": SIMPLE_PANDAS_DF,
         # Just some random data to test multiple keys, different sort
@@ -902,18 +894,18 @@ def test_pandas_set_sort_columns(de_fixture: DataExplorerFixture):
 
         expected_df = df.sort_values(**expected_params)
 
-        de_fixture.check_sort_case(df, wrapped_keys, expected_df)
+        dxf.check_sort_case(df, wrapped_keys, expected_df)
 
         for filter_f, filters in filter_cases.get(df_name, []):
             expected_filtered = filter_f(df).sort_values(**expected_params)
-            de_fixture.check_sort_case(df, wrapped_keys, expected_filtered, filters=filters)
+            dxf.check_sort_case(df, wrapped_keys, expected_filtered, filters=filters)
 
 
 def test_pandas_change_schema_after_sort(
     shell: PositronShell,
     de_service: DataExplorerService,
     variables_comm: DummyComm,
-    de_fixture: DataExplorerFixture,
+    dxf: DataExplorerFixture,
 ):
     df = pd.DataFrame(
         {
@@ -929,26 +921,32 @@ def test_pandas_change_schema_after_sort(
 
     # Sort a column that is out of bounds for the table after the
     # schema change below
-    de_fixture.set_sort_columns("df", [{"column_index": 4, "ascending": True}])
+    dxf.set_sort_columns("df", [{"column_index": 4, "ascending": True}])
 
     expected_df = df[["a", "b"]]
-    de_fixture.register_table("expected_df", df)
+    dxf.register_table("expected_df", df)
 
     # Sort last column, and we will then change the schema
     shell.run_cell("df = df[['a', 'b']]")
     _check_update_variable(de_service, "df", update_type="schema", discard_state=True)
 
     # Call get_data_values and make sure it works
-    de_fixture.compare_tables("df", "expected_df", expected_df.shape)
+    dxf.compare_tables("df", "expected_df", expected_df.shape)
 
 
 def _profile_request(column_index, profile_type):
     return {"column_index": column_index, "type": profile_type}
 
 
-def test_pandas_profile_null_counts(de_fixture: DataExplorerFixture):
-    dxf = de_fixture
+def _get_null_count(column_index):
+    return _profile_request(column_index, "null_count")
+
+
+def _get_summary_stats(column_index):
+    return _profile_request(column_index, "summary_stats")
+
 
+def test_pandas_profile_null_counts(dxf: DataExplorerFixture):
     df1 = pd.DataFrame(
         {
             "a": [0, np.nan, 2, np.nan, 4, 5, 6],
@@ -962,26 +960,28 @@ def test_pandas_profile_null_counts(de_fixture: DataExplorerFixture):
     for name, df in tables.items():
         dxf.register_table(name, df)
 
-    def _null_count(column_index):
-        return _profile_request(column_index, "null_count")
-
     # tuples like (table_name, [ColumnProfileRequest], [results])
     all_profiles = [
-        _null_count(0),
-        _null_count(1),
-        _null_count(2),
-        _null_count(3),
+        _get_null_count(0),
+        _get_null_count(1),
+        _get_null_count(2),
+        _get_null_count(3),
     ]
     cases = [
         ("df1", [], []),
         (
             "df1",
-            [_null_count(3)],
+            [_get_null_count(3)],
             [0],
         ),
         (
             "df1",
-            [_null_count(0), _null_count(1), _null_count(2), _null_count(3)],
+            [
+                _get_null_count(0),
+                _get_null_count(1),
+                _get_null_count(2),
+                _get_null_count(3),
+            ],
             [2, 3, 4, 0],
         ),
     ]
@@ -1010,6 +1010,105 @@ def _null_count(column_index):
         assert results == ex_results
 
 
+EPSILON = 1e-7
+
+
+def _assert_close(expected, actual):
+    assert np.abs(actual - expected) < EPSILON
+
+
+def _assert_numeric_stats_equal(expected, actual):
+    for attr, value in expected.items():
+        _assert_close(value, float(actual.get(attr)))
+
+
+def _assert_string_stats_equal(expected, actual):
+    assert expected["num_empty"] == actual["num_empty"]
+    assert expected["num_unique"] == actual["num_unique"]
+
+
+def _assert_boolean_stats_equal(expected, actual):
+    assert expected["true_count"] == actual["true_count"]
+    assert expected["false_count"] == actual["false_count"]
+
+
+def test_pandas_profile_summary_stats(dxf: DataExplorerFixture):
+    arr = np.random.standard_normal(100)
+    arr_with_nulls = arr.copy()
+    arr_with_nulls[::10] = np.nan
+
+    df1 = pd.DataFrame(
+        {
+            "a": arr,
+            "b": arr_with_nulls,
+            "c": [False, False, False, True, None] * 20,
+            "d": [
+                "foo",
+                "",
+                "baz",
+                "qux",
+                "foo",
+                None,
+                "bar",
+                "",
+                "bar",
+                "zzz",
+            ]
+            * 10,
+        }
+    )
+    dxf.register_table("df1", df1)
+
+    cases = [
+        (
+            "df1",
+            0,
+            {
+                "min_value": arr.min(),
+                "max_value": arr.max(),
+                "mean": df1["a"].mean(),
+                "stdev": df1["a"].std(),
+                "median": df1["a"].median(),
+            },
+        ),
+        (
+            "df1",
+            1,
+            {
+                "min_value": df1["b"].min(),
+                "max_value": df1["b"].max(),
+                "mean": df1["b"].mean(),
+                "stdev": df1["b"].std(),
+                "median": df1["b"].median(),
+            },
+        ),
+        (
+            "df1",
+            2,
+            {"true_count": 20, "false_count": 60},
+        ),
+        (
+            "df1",
+            3,
+            {"num_empty": 20, "num_unique": 6},
+        ),
+    ]
+
+    for table_name, col_index, ex_result in cases:
+        profiles = [_get_summary_stats(col_index)]
+        results = dxf.get_column_profiles(table_name, profiles)
+
+        stats = results[0]["summary_stats"]
+        ui_type = stats["type_display"]
+
+        if ui_type == ColumnDisplayType.Number:
+            _assert_numeric_stats_equal(ex_result, stats["number_stats"])
+        elif ui_type == ColumnDisplayType.String:
+            _assert_string_stats_equal(ex_result, stats["string_stats"])
+        elif ui_type == ColumnDisplayType.Boolean:
+            _assert_boolean_stats_equal(ex_result, stats["boolean_stats"])
+
+
 # ----------------------------------------------------------------------
 # Test RPCs for polars DataFrame
 
diff --git a/positron/comms/data_explorer-backend-openrpc.json b/positron/comms/data_explorer-backend-openrpc.json
index 09eb728a433..91d033fb13b 100644
--- a/positron/comms/data_explorer-backend-openrpc.json
+++ b/positron/comms/data_explorer-backend-openrpc.json
@@ -289,6 +289,22 @@
 	"components": {
 		"contentDescriptors": {},
 		"schemas": {
+			"column_display_type": {
+				"name": "column_display_type",
+				"type": "string",
+				"description": "Canonical Positron display name of data type",
+				"enum": [
+					"number",
+					"boolean",
+					"string",
+					"date",
+					"datetime",
+					"time",
+					"array",
+					"struct",
+					"unknown"
+				]
+			},
 			"column_schema": {
 				"type": "object",
 				"description": "Schema for a column in a table",
@@ -312,19 +328,8 @@
 						"description": "Exact name of data type used by underlying table"
 					},
 					"type_display": {
-						"type": "string",
 						"description": "Canonical Positron display name of data type",
-						"enum": [
-							"number",
-							"boolean",
-							"string",
-							"date",
-							"datetime",
-							"time",
-							"array",
-							"struct",
-							"unknown"
-						]
+						"$ref": "#/components/schemas/column_display_type"
 					},
 					"description": {
 						"type": "string",
@@ -560,10 +565,38 @@
 				}
 			},
 			"column_summary_stats": {
+				"type": "object",
+				"description": "Profile result containing summary stats for a column based on the data type",
+				"required": [
+					"type_display"
+				],
+				"properties": {
+					"type_display": {
+						"description": "Canonical Positron display name of data type",
+						"$ref": "#/components/schemas/column_display_type"
+					},
+					"number_stats": {
+						"description": "Statistics for a numeric data type",
+						"$ref": "#/components/schemas/summary_stats_number"
+					},
+					"string_stats": {
+						"description": "Statistics for a string-like data type",
+						"$ref": "#/components/schemas/summary_stats_string"
+					},
+					"boolean_stats": {
+						"description": "Statistics for a boolean data type",
+						"$ref": "#/components/schemas/summary_stats_boolean"
+					}
+				}
+			},
+			"summary_stats_number": {
 				"type": "object",
 				"required": [
 					"min_value",
-					"max_value"
+					"max_value",
+					"mean",
+					"median",
+					"stdev"
 				],
 				"properties": {
 					"min_value": {
@@ -574,7 +607,7 @@
 						"type": "string",
 						"description": "Maximum value as string"
 					},
-					"mean_value": {
+					"mean": {
 						"type": "string",
 						"description": "Average value as string"
 					},
@@ -582,13 +615,43 @@
 						"type": "string",
 						"description": "Sample median (50% value) value as string"
 					},
-					"q25": {
+					"stdev": {
 						"type": "string",
-						"description": "25th percentile value as string"
+						"description": "Sample standard deviation as a string"
+					}
+				}
+			},
+			"summary_stats_boolean": {
+				"type": "object",
+				"required": [
+					"true_count",
+					"false_count"
+				],
+				"properties": {
+					"true_count": {
+						"type": "integer",
+						"description": "The number of non-null true values"
 					},
-					"q75": {
-						"type": "string",
-						"description": "75th percentile value as string"
+					"false_count": {
+						"type": "integer",
+						"description": "The number of non-null false values"
+					}
+				}
+			},
+			"summary_stats_string": {
+				"type": "object",
+				"required": [
+					"num_empty",
+					"num_unique"
+				],
+				"properties": {
+					"num_empty": {
+						"type": "integer",
+						"description": "The number of empty / length-zero values"
+					},
+					"num_unique": {
+						"type": "integer",
+						"description": "The exact number of distinct values"
 					}
 				}
 			},
diff --git a/positron/comms/generate-comms.ts b/positron/comms/generate-comms.ts
index f06dfe97910..c98530a95d0 100644
--- a/positron/comms/generate-comms.ts
+++ b/positron/comms/generate-comms.ts
@@ -212,10 +212,12 @@ function deriveType(contracts: Array<any>,
 		}
 	} else if (schema.type === 'string' && schema.enum) {
 		if (context.length < 2) {
-			throw new Error(`missing context (need at least 2 elements): ${context[0]}`);
+			return snakeCaseToSentenceCase(context[0]);
+		} else {
+			// An enum field within another type, we add the context prefix
+			return snakeCaseToSentenceCase(context[1]) +
+				snakeCaseToSentenceCase(context[0]);
 		}
-		return snakeCaseToSentenceCase(context[1]) +
-			snakeCaseToSentenceCase(context[0]);
 	} else {
 		if (Object.keys(typeMap).includes(schema.type)) {
 			return typeMap[schema.type];
@@ -301,7 +303,7 @@ function* enumVisitor(
 				// and recurse
 				yield* enumVisitor(
 					[contract['name'], ...context], contract[key], callback);
-			} else if (key === 'properties' || key === 'params') {
+			} else if (key === 'properties' || key === 'params' || key === 'schemas' || key === 'components') {
 				// If this is a properties or params object, recurse into each
 				// property, but don't push the parent name onto the context
 				yield* enumVisitor(
@@ -397,21 +399,6 @@ use serde::Serialize;
 			continue;
 		}
 
-		// Create type aliases for all the shared types
-		if (source.components && source.components.schemas) {
-			for (const key of Object.keys(source.components.schemas)) {
-				const schema = source.components.schemas[key];
-				if (schema.type !== 'object') {
-					yield formatComment('/// ', schema.description);
-					yield `type ${snakeCaseToSentenceCase(key)} = `;
-					yield deriveType(contracts, RustTypeMap,
-						[schema.name ? schema.name : key],
-						schema);
-					yield ';\n\n';
-				}
-			}
-		}
-
 		// Create structs for all object types
 		yield* objectVisitor([], source, function* (context: Array<string>, o: Record<string, any>) {
 			if (o.description) {
@@ -463,12 +450,22 @@ use serde::Serialize;
 
 		// Create enums for all enum types
 		yield* enumVisitor([], source, function* (context: Array<string>, values: Array<string>) {
-			yield formatComment(`/// `,
-				`Possible values for ` +
-				snakeCaseToSentenceCase(context[0]) + ` in ` +
-				snakeCaseToSentenceCase(context[1]));
-			yield '#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]\n';
-			yield `pub enum ${snakeCaseToSentenceCase(context[1])}${snakeCaseToSentenceCase(context[0])} {\n`;
+			if (context.length === 1) {
+				// Shared enum at the components.schemas level
+				yield formatComment(`/// `,
+					`Possible values for ` +
+					snakeCaseToSentenceCase(context[0]));
+				yield '#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]\n';
+				yield `pub enum ${snakeCaseToSentenceCase(context[0])} {\n`;
+			} else {
+				// Enum field within another interface
+				yield formatComment(`/// `,
+					`Possible values for ` +
+					snakeCaseToSentenceCase(context[0]) + ` in ` +
+					snakeCaseToSentenceCase(context[1]));
+				yield '#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]\n';
+				yield `pub enum ${snakeCaseToSentenceCase(context[1])}${snakeCaseToSentenceCase(context[0])} {\n`;
+			}
 			for (let i = 0; i < values.length; i++) {
 				const value = values[i];
 				yield `\t#[serde(rename = "${value}")]\n`;
@@ -712,32 +709,28 @@ from ._vendor.pydantic import BaseModel, Field
 			continue;
 		}
 
-		// Create type aliases for all the shared types
-		if (source.components && source.components.schemas) {
-			for (const key of Object.keys(source.components.schemas)) {
-				const schema = source.components.schemas[key];
-				if (schema.type !== 'object') {
-					yield formatComment('# ', schema.description);
-					yield `${snakeCaseToSentenceCase(key)} = `;
-					yield deriveType(contracts, PythonTypeMap,
-						[schema.name ? schema.name : key],
-						schema);
-					yield '\n\n';
-				}
-			}
-		}
-
 		// Create enums for all enum types
 		yield* enumVisitor([], source, function* (context: Array<string>, values: Array<string>) {
-			yield '@enum.unique\n';
-			yield `class ${snakeCaseToSentenceCase(context[1])}`;
-			yield `${snakeCaseToSentenceCase(context[0])}(str, enum.Enum):\n`;
-			yield '    """\n';
-			yield formatComment(`    `,
-				`Possible values for ` +
-				snakeCaseToSentenceCase(context[0]) +
-				` in ` +
-				snakeCaseToSentenceCase(context[1]));
+			if (context.length === 1) {
+				// Shared enum at the components.schemas level
+				yield '@enum.unique\n';
+				yield `class ${snakeCaseToSentenceCase(context[0])}(str, enum.Enum):\n`;
+				yield '    """\n';
+				yield formatComment(`    `,
+					`Possible values for ` +
+					snakeCaseToSentenceCase(context[0]));
+			} else {
+				// Enum field within another interface
+				yield '@enum.unique\n';
+				yield `class ${snakeCaseToSentenceCase(context[1])}`;
+				yield `${snakeCaseToSentenceCase(context[0])}(str, enum.Enum):\n`;
+				yield '    """\n';
+				yield formatComment(`    `,
+					`Possible values for ` +
+					snakeCaseToSentenceCase(context[0]) +
+					` in ` +
+					snakeCaseToSentenceCase(context[1]));
+			}
 			yield '    """\n';
 			yield '\n';
 			for (let i = 0; i < values.length; i++) {
@@ -1051,12 +1044,22 @@ import { IRuntimeClientInstance } from 'vs/workbench/services/languageRuntime/co
 		// Create enums for all enum types
 		yield* enumVisitor([], source, function* (context: Array<string>, values: Array<string>) {
 			yield '/**\n';
-			yield formatComment(` * `,
-				`Possible values for ` +
-				snakeCaseToSentenceCase(context[0]) + ` in ` +
-				snakeCaseToSentenceCase(context[1]));
-			yield ' */\n';
-			yield `export enum ${snakeCaseToSentenceCase(context[1])}${snakeCaseToSentenceCase(context[0])} {\n`;
+			if (context.length === 1) {
+				// Shared enum at the components.schemas level
+				yield formatComment(` * `,
+					`Possible values for ` +
+					snakeCaseToSentenceCase(context[0]));
+				yield ' */\n';
+				yield `export enum ${snakeCaseToSentenceCase(context[0])} {\n`;
+			} else {
+				// Enum field within another interface
+				yield formatComment(` * `,
+					`Possible values for ` +
+					snakeCaseToSentenceCase(context[0]) + ` in ` +
+					snakeCaseToSentenceCase(context[1]));
+				yield ' */\n';
+				yield `export enum ${snakeCaseToSentenceCase(context[1])}${snakeCaseToSentenceCase(context[0])} {\n`;
+			}
 			for (let i = 0; i < values.length; i++) {
 				const value = values[i];
 				yield `\t${snakeCaseToSentenceCase(value)} = '${value}'`;
@@ -1070,25 +1073,6 @@ import { IRuntimeClientInstance } from 'vs/workbench/services/languageRuntime/co
 		});
 	}
 
-	for (const source of [backend, frontend]) {
-		if (!source) {
-			continue;
-		}
-		if (source.components && source.components.schemas) {
-			for (const key of Object.keys(source.components.schemas)) {
-				const schema = source.components.schemas[key];
-				if (schema.type !== 'object') {
-					yield `/**\n`;
-					yield formatComment(' * ', schema.description);
-					yield ' */\n';
-					yield `export type ${snakeCaseToSentenceCase(key)} = `;
-					yield deriveType(contracts, TypescriptTypeMap, [key], schema);
-					yield ';\n\n';
-				}
-			}
-		}
-	}
-
 	if (frontend) {
 		const events: string[] = [];
 		const requests: string[] = [];
diff --git a/src/vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/components/addEditRowFilterModalPopup/addEditRowFilterModalPopup.tsx b/src/vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/components/addEditRowFilterModalPopup/addEditRowFilterModalPopup.tsx
index 58cb5feb0a7..41980f2fd13 100644
--- a/src/vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/components/addEditRowFilterModalPopup/addEditRowFilterModalPopup.tsx
+++ b/src/vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/components/addEditRowFilterModalPopup/addEditRowFilterModalPopup.tsx
@@ -18,7 +18,7 @@ import { PositronModalPopup } from 'vs/workbench/browser/positronComponents/posi
 import { PositronModalReactRenderer } from 'vs/workbench/browser/positronModalReactRenderer/positronModalReactRenderer';
 import { DropDownListBoxSeparator } from 'vs/workbench/browser/positronComponents/dropDownListBox/dropDownListBoxSeparator';
 import { DataExplorerClientInstance } from 'vs/workbench/services/languageRuntime/common/languageRuntimeDataExplorerClient';
-import { ColumnSchema, ColumnSchemaTypeDisplay } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
+import { ColumnSchema, ColumnDisplayType } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
 import { RowFilterParameter } from 'vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/components/addEditRowFilterModalPopup/components/rowFilterParameter';
 import { DropDownColumnSelector } from 'vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/components/addEditRowFilterModalPopup/components/dropDownColumnSelector';
 import { RangeRowFilter, RowFilter, RowFilterCondition, RowFilterIsBetween, RowFilterIsEmpty, RowFilterIsEqualTo, RowFilterIsGreaterThan, RowFilterIsLessThan, RowFilterIsNotBetween, RowFilterIsNotEmpty, SingleValueRowFilter } from 'vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/components/addEditRowFilterModalPopup/rowFilter';
@@ -51,21 +51,21 @@ const validateRowFilterValue = (columnSchema: ColumnSchema, value: string) => {
 	// Validate the row filter value that was supplied based on the column schema type.
 	switch (columnSchema.type_display) {
 		// Number.
-		case ColumnSchemaTypeDisplay.Number:
+		case ColumnDisplayType.Number:
 			return isNumber();
 
 		// Boolean.
-		case ColumnSchemaTypeDisplay.Boolean:
+		case ColumnDisplayType.Boolean:
 			return isBoolean();
 
 		// String.
-		case ColumnSchemaTypeDisplay.String:
+		case ColumnDisplayType.String:
 			return true;
 
 		// TODO: Add more complete validation.
-		case ColumnSchemaTypeDisplay.Date:
-		case ColumnSchemaTypeDisplay.Datetime:
-		case ColumnSchemaTypeDisplay.Time:
+		case ColumnDisplayType.Date:
+		case ColumnDisplayType.Datetime:
+		case ColumnDisplayType.Time:
 			return isDate();
 
 		// Can't get here.
@@ -160,10 +160,10 @@ export const AddEditRowFilterModalPopup = (props: AddEditRowFilterModalPopupProp
 
 		// Add is less than / is greater than conditions.
 		switch (selectedColumnSchema.type_display) {
-			case ColumnSchemaTypeDisplay.Number:
-			case ColumnSchemaTypeDisplay.Date:
-			case ColumnSchemaTypeDisplay.Datetime:
-			case ColumnSchemaTypeDisplay.Time:
+			case ColumnDisplayType.Number:
+			case ColumnDisplayType.Date:
+			case ColumnDisplayType.Datetime:
+			case ColumnDisplayType.Time:
 				conditionEntries.push(new DropDownListBoxItem({
 					identifier: RowFilterCondition.CONDITION_IS_LESS_THAN,
 					title: localize(
@@ -183,12 +183,12 @@ export const AddEditRowFilterModalPopup = (props: AddEditRowFilterModalPopupProp
 
 		// Add is equal to condition.
 		switch (selectedColumnSchema.type_display) {
-			case ColumnSchemaTypeDisplay.Number:
-			case ColumnSchemaTypeDisplay.Boolean:
-			case ColumnSchemaTypeDisplay.String:
-			case ColumnSchemaTypeDisplay.Date:
-			case ColumnSchemaTypeDisplay.Datetime:
-			case ColumnSchemaTypeDisplay.Time:
+			case ColumnDisplayType.Number:
+			case ColumnDisplayType.Boolean:
+			case ColumnDisplayType.String:
+			case ColumnDisplayType.Date:
+			case ColumnDisplayType.Datetime:
+			case ColumnDisplayType.Time:
 				conditionEntries.push(new DropDownListBoxItem({
 					identifier: RowFilterCondition.CONDITION_IS_EQUAL_TO,
 					title: localize(
@@ -201,10 +201,10 @@ export const AddEditRowFilterModalPopup = (props: AddEditRowFilterModalPopupProp
 
 		// Add is between / is not between conditions.
 		switch (selectedColumnSchema.type_display) {
-			case ColumnSchemaTypeDisplay.Number:
-			case ColumnSchemaTypeDisplay.Date:
-			case ColumnSchemaTypeDisplay.Datetime:
-			case ColumnSchemaTypeDisplay.Time:
+			case ColumnDisplayType.Number:
+			case ColumnDisplayType.Date:
+			case ColumnDisplayType.Datetime:
+			case ColumnDisplayType.Time:
 				conditionEntries.push(new DropDownListBoxSeparator());
 				conditionEntries.push(new DropDownListBoxItem({
 					identifier: RowFilterCondition.CONDITION_IS_BETWEEN,
diff --git a/src/vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/utility/columnSchemaUtilities.ts b/src/vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/utility/columnSchemaUtilities.ts
index 2446d3dce22..5fe5c9f9b5c 100644
--- a/src/vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/utility/columnSchemaUtilities.ts
+++ b/src/vs/workbench/browser/positronDataExplorer/components/dataExplorerPanel/utility/columnSchemaUtilities.ts
@@ -2,7 +2,7 @@
  *  Copyright (C) 2024 Posit Software, PBC. All rights reserved.
  *--------------------------------------------------------------------------------------------*/
 
-import { ColumnSchema, ColumnSchemaTypeDisplay } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
+import { ColumnSchema, ColumnDisplayType } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
 
 /**
  * Returns the data type icon for the column schema.
@@ -16,31 +16,31 @@ export const columnSchemaDataTypeIcon = (columnSchema?: ColumnSchema) => {
 
 	// Determine the alignment based on type.
 	switch (columnSchema.type_display) {
-		case ColumnSchemaTypeDisplay.Number:
+		case ColumnDisplayType.Number:
 			return 'codicon-positron-data-type-number';
 
-		case ColumnSchemaTypeDisplay.Boolean:
+		case ColumnDisplayType.Boolean:
 			return 'codicon-positron-data-type-boolean';
 
-		case ColumnSchemaTypeDisplay.String:
+		case ColumnDisplayType.String:
 			return 'codicon-positron-data-type-string';
 
-		case ColumnSchemaTypeDisplay.Date:
+		case ColumnDisplayType.Date:
 			return 'codicon-positron-data-type-date';
 
-		case ColumnSchemaTypeDisplay.Datetime:
+		case ColumnDisplayType.Datetime:
 			return 'codicon-positron-data-type-date-time';
 
-		case ColumnSchemaTypeDisplay.Time:
+		case ColumnDisplayType.Time:
 			return 'codicon-positron-data-type-time';
 
-		case ColumnSchemaTypeDisplay.Array:
+		case ColumnDisplayType.Array:
 			return 'codicon-positron-data-type-array';
 
-		case ColumnSchemaTypeDisplay.Struct:
+		case ColumnDisplayType.Struct:
 			return 'codicon-positron-data-type-struct';
 
-		case ColumnSchemaTypeDisplay.Unknown:
+		case ColumnDisplayType.Unknown:
 			return 'codicon-positron-data-type-unknown';
 
 		// This shouldn't ever happen.
diff --git a/src/vs/workbench/services/languageRuntime/common/positronDataExplorerComm.ts b/src/vs/workbench/services/languageRuntime/common/positronDataExplorerComm.ts
index 8af3e7e49be..ae4f15dc01d 100644
--- a/src/vs/workbench/services/languageRuntime/common/positronDataExplorerComm.ts
+++ b/src/vs/workbench/services/languageRuntime/common/positronDataExplorerComm.ts
@@ -112,7 +112,7 @@ export interface ColumnSchema {
 	/**
 	 * Canonical Positron display name of data type
 	 */
-	type_display: ColumnSchemaTypeDisplay;
+	type_display: ColumnDisplayType;
 
 	/**
 	 * Column annotation / description
@@ -310,9 +310,36 @@ export interface ColumnProfileResult {
 }
 
 /**
- * ColumnSummaryStats in Schemas
+ * Profile result containing summary stats for a column based on the data
+ * type
  */
 export interface ColumnSummaryStats {
+	/**
+	 * Canonical Positron display name of data type
+	 */
+	type_display: ColumnDisplayType;
+
+	/**
+	 * Statistics for a numeric data type
+	 */
+	number_stats?: SummaryStatsNumber;
+
+	/**
+	 * Statistics for a string-like data type
+	 */
+	string_stats?: SummaryStatsString;
+
+	/**
+	 * Statistics for a boolean data type
+	 */
+	boolean_stats?: SummaryStatsBoolean;
+
+}
+
+/**
+ * SummaryStatsNumber in Schemas
+ */
+export interface SummaryStatsNumber {
 	/**
 	 * Minimum value as string
 	 */
@@ -326,22 +353,49 @@ export interface ColumnSummaryStats {
 	/**
 	 * Average value as string
 	 */
-	mean_value?: string;
+	mean: string;
 
 	/**
 	 * Sample median (50% value) value as string
 	 */
-	median?: string;
+	median: string;
 
 	/**
-	 * 25th percentile value as string
+	 * Sample standard deviation as a string
+	 */
+	stdev: string;
+
+}
+
+/**
+ * SummaryStatsBoolean in Schemas
+ */
+export interface SummaryStatsBoolean {
+	/**
+	 * The number of non-null true values
+	 */
+	true_count: number;
+
+	/**
+	 * The number of non-null false values
+	 */
+	false_count: number;
+
+}
+
+/**
+ * SummaryStatsString in Schemas
+ */
+export interface SummaryStatsString {
+	/**
+	 * The number of empty / length-zero values
 	 */
-	q25?: string;
+	num_empty: number;
 
 	/**
-	 * 75th percentile value as string
+	 * The exact number of distinct values
 	 */
-	q75?: string;
+	num_unique: number;
 
 }
 
@@ -432,9 +486,9 @@ export interface ColumnSortKey {
 }
 
 /**
- * Possible values for TypeDisplay in ColumnSchema
+ * Possible values for ColumnDisplayType
  */
-export enum ColumnSchemaTypeDisplay {
+export enum ColumnDisplayType {
 	Number = 'number',
 	Boolean = 'boolean',
 	String = 'string',
@@ -540,7 +594,7 @@ export class PositronDataExplorerComm extends PositronBaseComm {
 	 *
 	 * Search schema for column names matching a passed substring
 	 *
-	 * @param searchTerm Substring to match for (currently case insensitive
+	 * @param searchTerm Substring to match for (currently case insensitive)
 	 * @param startIndex Index (starting from zero) of first result to fetch
 	 * @param maxResults Maximum number of resulting column schemas to fetch
 	 * from the start index
diff --git a/src/vs/workbench/services/positronDataExplorer/browser/components/columnSummaryCell.tsx b/src/vs/workbench/services/positronDataExplorer/browser/components/columnSummaryCell.tsx
index 691003a554b..1c49daf3e55 100644
--- a/src/vs/workbench/services/positronDataExplorer/browser/components/columnSummaryCell.tsx
+++ b/src/vs/workbench/services/positronDataExplorer/browser/components/columnSummaryCell.tsx
@@ -11,7 +11,7 @@ import * as React from 'react';
 // Other dependencies.
 import { ProfileNumber } from 'vs/workbench/services/positronDataExplorer/browser/components/profileNumber';
 import { ProfileString } from 'vs/workbench/services/positronDataExplorer/browser/components/profileString';
-import { ColumnSchema, ColumnSchemaTypeDisplay } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
+import { ColumnSchema, ColumnDisplayType } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
 import { TableSummaryDataGridInstance } from 'vs/workbench/services/positronDataExplorer/browser/tableSummaryDataGridInstance';
 
 /**
@@ -37,31 +37,31 @@ export const ColumnSummaryCell = (props: ColumnSummaryCellProps) => {
 	const dataTypeIcon = () => {
 		// Determine the alignment based on type.
 		switch (props.columnSchema.type_display) {
-			case ColumnSchemaTypeDisplay.Number:
+			case ColumnDisplayType.Number:
 				return 'codicon-positron-data-type-number';
 
-			case ColumnSchemaTypeDisplay.Boolean:
+			case ColumnDisplayType.Boolean:
 				return 'codicon-positron-data-type-boolean';
 
-			case ColumnSchemaTypeDisplay.String:
+			case ColumnDisplayType.String:
 				return 'codicon-positron-data-type-string';
 
-			case ColumnSchemaTypeDisplay.Date:
+			case ColumnDisplayType.Date:
 				return 'codicon-positron-data-type-date';
 
-			case ColumnSchemaTypeDisplay.Datetime:
+			case ColumnDisplayType.Datetime:
 				return 'codicon-positron-data-type-date-time';
 
-			case ColumnSchemaTypeDisplay.Time:
+			case ColumnDisplayType.Time:
 				return 'codicon-positron-data-type-time';
 
-			case ColumnSchemaTypeDisplay.Array:
+			case ColumnDisplayType.Array:
 				return 'codicon-positron-data-type-array';
 
-			case ColumnSchemaTypeDisplay.Struct:
+			case ColumnDisplayType.Struct:
 				return 'codicon-positron-data-type-struct';
 
-			case ColumnSchemaTypeDisplay.Unknown:
+			case ColumnDisplayType.Unknown:
 				return 'codicon-positron-data-type-unknown';
 
 			// This shouldn't ever happen.
@@ -75,33 +75,38 @@ export const ColumnSummaryCell = (props: ColumnSummaryCellProps) => {
 	 * @returns The profile component.
 	 */
 	const profile = () => {
+		// Hack just to get things working
+		props.instance.computeColumnSummaryStats(props.columnIndex);
 		// Determine the alignment based on type.
 		switch (props.columnSchema.type_display) {
-			case ColumnSchemaTypeDisplay.Number:
-				return <ProfileNumber />;
+			case ColumnDisplayType.Number:
+				return <ProfileNumber
+					instance={props.instance}
+					columnIndex={props.columnIndex}
+				/>;
 
-			case ColumnSchemaTypeDisplay.Boolean:
+			case ColumnDisplayType.Boolean:
 				return null;
 
-			case ColumnSchemaTypeDisplay.String:
+			case ColumnDisplayType.String:
 				return <ProfileString />;
 
-			case ColumnSchemaTypeDisplay.Date:
+			case ColumnDisplayType.Date:
 				return null;
 
-			case ColumnSchemaTypeDisplay.Datetime:
+			case ColumnDisplayType.Datetime:
 				return null;
 
-			case ColumnSchemaTypeDisplay.Time:
+			case ColumnDisplayType.Time:
 				return null;
 
-			case ColumnSchemaTypeDisplay.Array:
+			case ColumnDisplayType.Array:
 				return null;
 
-			case ColumnSchemaTypeDisplay.Struct:
+			case ColumnDisplayType.Struct:
 				return null;
 
-			case ColumnSchemaTypeDisplay.Unknown:
+			case ColumnDisplayType.Unknown:
 				return null;
 
 			// This shouldn't ever happen.
diff --git a/src/vs/workbench/services/positronDataExplorer/browser/components/profileNumber.tsx b/src/vs/workbench/services/positronDataExplorer/browser/components/profileNumber.tsx
index 679e9a7cdbe..c726fd20368 100644
--- a/src/vs/workbench/services/positronDataExplorer/browser/components/profileNumber.tsx
+++ b/src/vs/workbench/services/positronDataExplorer/browser/components/profileNumber.tsx
@@ -8,10 +8,14 @@ import 'vs/css!./profileNumber';
 // React.
 import * as React from 'react';
 
+import { TableSummaryDataGridInstance } from 'vs/workbench/services/positronDataExplorer/browser/tableSummaryDataGridInstance';
+
 /**
  * ProfileNumberProps interface.
  */
 interface ProfileNumberProps {
+	instance: TableSummaryDataGridInstance;
+	columnIndex: number;
 }
 
 /**
@@ -20,33 +24,38 @@ interface ProfileNumberProps {
  * @returns The rendered component.
  */
 export const ProfileNumber = (props: ProfileNumberProps) => {
+	// Hack
+	let stats: any = props.instance.getColumnSummaryStats(props.columnIndex)?.number_stats!;
+	if (!stats) {
+		stats = {};
+	}
 	return (
 		<div className='tabular-info'>
 			<div className='labels'>
 				<div className='label'>NA</div>
-				<div className='label'>Median</div>
 				<div className='label'>Mean</div>
+				<div className='label'>Median</div>
 				<div className='label'>SD</div>
 				<div className='label'>Min</div>
 				<div className='label'>Max</div>
 			</div>
 			<div className='values'>
 				<div className='values-left'>
-					<div className='value'>12</div>
-					<div className='value'>1</div>
-					<div className='value'>4</div>
-					<div className='value'>2</div>
-					<div className='value'>5</div>
-					<div className='value'>102</div>
+					<div className='value'>-999999</div>
+					<div className='value'>{stats.mean}</div>
+					<div className='value'>{stats.median}</div>
+					<div className='value'>{stats.stdev}</div>
+					<div className='value'>{stats.min_value}</div>
+					<div className='value'>{stats.max_value}</div>
 				</div>
-				<div className='values-right'>
+				{/* <div className='values-right'>
 					<div className='value'>&nbsp;</div>
 					<div className='value'>.51</div>
 					<div className='value'>.20</div>
 					<div className='value'>.24</div>
 					<div className='value'>&nbsp;</div>
 					<div className='value'>.44</div>
-				</div>
+				</div> */}
 			</div>
 		</div>
 	);
diff --git a/src/vs/workbench/services/positronDataExplorer/browser/positronDataExplorerColumn.ts b/src/vs/workbench/services/positronDataExplorer/browser/positronDataExplorerColumn.ts
index ac21df778cf..2eb6dc74e9b 100644
--- a/src/vs/workbench/services/positronDataExplorer/browser/positronDataExplorerColumn.ts
+++ b/src/vs/workbench/services/positronDataExplorer/browser/positronDataExplorerColumn.ts
@@ -3,7 +3,7 @@
  *--------------------------------------------------------------------------------------------*/
 
 import { DataColumnAlignment } from 'vs/workbench/browser/positronDataGrid/interfaces/dataColumn';
-import { ColumnSchema, ColumnSchemaTypeDisplay } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
+import { ColumnSchema, ColumnDisplayType } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
 import { IPositronDataExplorerColumn } from 'vs/workbench/services/positronDataExplorer/browser/interfaces/positronDataExplorerColumn';
 
 /**
@@ -65,31 +65,31 @@ export class PositronDataExplorerColumn implements IPositronDataExplorerColumn {
 	get alignment() {
 		// Determine the alignment based on type.
 		switch (this.columnSchema.type_display) {
-			case ColumnSchemaTypeDisplay.Number:
+			case ColumnDisplayType.Number:
 				return DataColumnAlignment.Right;
 
-			case ColumnSchemaTypeDisplay.Boolean:
+			case ColumnDisplayType.Boolean:
 				return DataColumnAlignment.Left;
 
-			case ColumnSchemaTypeDisplay.String:
+			case ColumnDisplayType.String:
 				return DataColumnAlignment.Left;
 
-			case ColumnSchemaTypeDisplay.Date:
+			case ColumnDisplayType.Date:
 				return DataColumnAlignment.Right;
 
-			case ColumnSchemaTypeDisplay.Datetime:
+			case ColumnDisplayType.Datetime:
 				return DataColumnAlignment.Right;
 
-			case ColumnSchemaTypeDisplay.Time:
+			case ColumnDisplayType.Time:
 				return DataColumnAlignment.Right;
 
-			case ColumnSchemaTypeDisplay.Array:
+			case ColumnDisplayType.Array:
 				return DataColumnAlignment.Left;
 
-			case ColumnSchemaTypeDisplay.Struct:
+			case ColumnDisplayType.Struct:
 				return DataColumnAlignment.Left;
 
-			case ColumnSchemaTypeDisplay.Unknown:
+			case ColumnDisplayType.Unknown:
 				return DataColumnAlignment.Left;
 		}
 	}
diff --git a/src/vs/workbench/services/positronDataExplorer/browser/tableSummaryDataGridInstance.tsx b/src/vs/workbench/services/positronDataExplorer/browser/tableSummaryDataGridInstance.tsx
index 502957a3d54..8e7d379c53a 100644
--- a/src/vs/workbench/services/positronDataExplorer/browser/tableSummaryDataGridInstance.tsx
+++ b/src/vs/workbench/services/positronDataExplorer/browser/tableSummaryDataGridInstance.tsx
@@ -9,7 +9,7 @@ import * as React from 'react';
 import { Emitter } from 'vs/base/common/event';
 import { DataGridInstance } from 'vs/workbench/browser/positronDataGrid/classes/dataGridInstance';
 import { DataExplorerCache } from 'vs/workbench/services/positronDataExplorer/common/dataExplorerCache';
-import { ColumnSchemaTypeDisplay } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
+import { ColumnDisplayType, ColumnSummaryStats } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
 import { ColumnSummaryCell } from 'vs/workbench/services/positronDataExplorer/browser/components/columnSummaryCell';
 import { DataExplorerClientInstance } from 'vs/workbench/services/languageRuntime/common/languageRuntimeDataExplorerClient';
 
@@ -161,31 +161,31 @@ export class TableSummaryDataGridInstance extends DataGridInstance {
 
 		// Return the row height.
 		switch (columnSchema.type_display) {
-			case ColumnSchemaTypeDisplay.Number:
+			case ColumnDisplayType.Number:
 				return rowHeight(6);
 
-			case ColumnSchemaTypeDisplay.Boolean:
+			case ColumnDisplayType.Boolean:
 				return rowHeight(3);
 
-			case ColumnSchemaTypeDisplay.String:
+			case ColumnDisplayType.String:
 				return rowHeight(3);
 
-			case ColumnSchemaTypeDisplay.Date:
+			case ColumnDisplayType.Date:
 				return rowHeight(7);
 
-			case ColumnSchemaTypeDisplay.Datetime:
+			case ColumnDisplayType.Datetime:
 				return rowHeight(7);
 
-			case ColumnSchemaTypeDisplay.Time:
+			case ColumnDisplayType.Time:
 				return rowHeight(7);
 
-			case ColumnSchemaTypeDisplay.Array:
+			case ColumnDisplayType.Array:
 				return rowHeight(2);
 
-			case ColumnSchemaTypeDisplay.Struct:
+			case ColumnDisplayType.Struct:
 				return rowHeight(2);
 
-			case ColumnSchemaTypeDisplay.Unknown:
+			case ColumnDisplayType.Unknown:
 				return rowHeight(2);
 
 			// This shouldn't ever happen.
@@ -230,6 +230,14 @@ export class TableSummaryDataGridInstance extends DataGridInstance {
 			nullCount * 100 / this._dataExplorerCache.rows);
 	}
 
+	getColumnSummaryStats(columnIndex: number): ColumnSummaryStats | undefined {
+		return this._dataExplorerCache.getColumnSummaryStats(columnIndex);
+	}
+
+	computeColumnSummaryStats(columnIndex: number) {
+		this._dataExplorerCache.updateColumnSummaryStats([columnIndex]);
+	}
+
 	//#endregion DataGridInstance Methods
 
 	//#region Public Events
diff --git a/src/vs/workbench/services/positronDataExplorer/common/dataExplorerCache.ts b/src/vs/workbench/services/positronDataExplorer/common/dataExplorerCache.ts
index f3c7961ab8d..1696850f065 100644
--- a/src/vs/workbench/services/positronDataExplorer/common/dataExplorerCache.ts
+++ b/src/vs/workbench/services/positronDataExplorer/common/dataExplorerCache.ts
@@ -4,7 +4,7 @@
 
 import { Emitter } from 'vs/base/common/event';
 import { Disposable } from 'vs/base/common/lifecycle';
-import { ColumnProfileRequestType, ColumnSchema, TableData } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
+import { ColumnProfileRequestType, ColumnSchema, ColumnSummaryStats, TableData } from 'vs/workbench/services/languageRuntime/common/positronDataExplorerComm';
 import { DataExplorerClientInstance } from 'vs/workbench/services/languageRuntime/common/languageRuntimeDataExplorerClient';
 
 /**
@@ -72,6 +72,11 @@ export class DataExplorerCache extends Disposable {
 	 */
 	private readonly _columnNullCountCache = new Map<number, number>();
 
+	/**
+	 * Gets the column summary stats cache.
+	 */
+	private readonly _columnSummaryStatsCache = new Map<number, ColumnSummaryStats>();
+
 	/**
 	 * Gets the row label cache.
 	 */
@@ -107,6 +112,7 @@ export class DataExplorerCache extends Disposable {
 			// Clear the column schema cache, row label cache, and data cell cache.
 			this._columnSchemaCache.clear();
 			this._columnNullCountCache.clear();
+			this._columnSummaryStatsCache.clear();
 			this._rowLabelCache.clear();
 			this._dataCellCache.clear();
 		}));
@@ -117,6 +123,7 @@ export class DataExplorerCache extends Disposable {
 			this._rowLabelCache.clear();
 			this._dataCellCache.clear();
 			this._columnNullCountCache.clear();
+			this._columnSummaryStatsCache.clear();
 		}));
 	}
 
@@ -158,6 +165,7 @@ export class DataExplorerCache extends Disposable {
 		this._rowLabelCache.clear();
 		this._dataCellCache.clear();
 		this._columnNullCountCache.clear();
+		this._columnSummaryStatsCache.clear();
 	}
 
 	/**
@@ -190,6 +198,35 @@ export class DataExplorerCache extends Disposable {
 		return this._columnNullCountCache.get(columnIndex);
 	}
 
+	/**
+	 * Gets the cached summary stats for the specified column index.
+	 * @param columnIndex The column index.
+	 * @returns ColumnSummaryStats in the specified column index.
+	 */
+	getColumnSummaryStats(columnIndex: number) {
+		return this._columnSummaryStatsCache.get(columnIndex);
+	}
+
+	async updateColumnSummaryStats(columnIndices: Array<number>) {
+		// Request the profiles
+		const results = await this._dataExplorerClientInstance.getColumnProfiles(
+			columnIndices.map(column_index => {
+				return {
+					column_index,
+					type: ColumnProfileRequestType.SummaryStats
+				};
+			})
+		);
+
+		// Update the column schema cache, overwriting any entries we already have cached.
+		for (let i = 0; i < results.length; i++) {
+			const stats = results[i].summary_stats;
+			if (stats !== undefined) {
+				this._columnSummaryStatsCache.set(columnIndices[i], stats);
+			}
+		}
+	}
+
 	/**
 	 * Gets the row label for the specified row index.
 	 * @param rowIndex The row index.