diff --git a/crates/polars-core/src/frame/group_by/mod.rs b/crates/polars-core/src/frame/group_by/mod.rs
index 7d1d9e761504..2a5aa2c40e45 100644
--- a/crates/polars-core/src/frame/group_by/mod.rs
+++ b/crates/polars-core/src/frame/group_by/mod.rs
@@ -841,6 +841,17 @@ impl<'df> GroupBy<'df> {
         df.as_single_chunk_par();
         Ok(df)
     }
+
+    pub fn sliced(mut self, slice: Option<(i64, usize)>) -> Self {
+        match slice {
+            None => self,
+            Some((offset, length)) => {
+                self.groups = (*self.groups.slice(offset, length)).clone();
+                self.selected_keys = self.keys_sliced(slice);
+                self
+            },
+        }
+    }
 }
 
 unsafe fn take_df(df: &DataFrame, g: GroupsIndicator) -> DataFrame {
diff --git a/crates/polars-mem-engine/src/executors/group_by.rs b/crates/polars-mem-engine/src/executors/group_by.rs
index 437b7fb574aa..f7a501424ed9 100644
--- a/crates/polars-mem-engine/src/executors/group_by.rs
+++ b/crates/polars-mem-engine/src/executors/group_by.rs
@@ -67,7 +67,7 @@ pub(super) fn group_by_helper(
     let gb = df.group_by_with_series(keys, true, maintain_order)?;
 
     if let Some(f) = apply {
-        return gb.apply(move |df| f.call_udf(df));
+        return gb.sliced(slice).apply(move |df| f.call_udf(df));
     }
 
     let mut groups = gb.get_groups();
diff --git a/py-polars/tests/unit/operations/test_group_by.py b/py-polars/tests/unit/operations/test_group_by.py
index cff43b43274c..645b978214f0 100644
--- a/py-polars/tests/unit/operations/test_group_by.py
+++ b/py-polars/tests/unit/operations/test_group_by.py
@@ -1153,3 +1153,31 @@ def test_group_by_agg_19173() -> None:
     out = df.head(0).group_by("g").agg((pl.col.x - pl.col.x.sum() * pl.col.x) ** 2)
     assert out.to_dict(as_series=False) == {"g": [], "x": []}
     assert out.schema == pl.Schema([("g", pl.Int64), ("x", pl.List(pl.Float64))])
+
+
+def test_group_by_map_groups_slice_pushdown_20002() -> None:
+    schema = {
+        "a": pl.Int8,
+        "b": pl.UInt8,
+    }
+
+    df = (
+        pl.LazyFrame(
+            data={"a": [1, 2, 3, 4, 5], "b": [90, 80, 70, 60, 50]},
+            schema=schema,
+        )
+        .group_by("a", maintain_order=True)
+        .map_groups(lambda df: df * 2.0, schema=schema)
+        .head(3)
+        .collect()
+    )
+
+    assert_frame_equal(
+        df,
+        pl.DataFrame(
+            {
+                "a": [2.0, 4.0, 6.0],
+                "b": [180.0, 160.0, 140.0],
+            }
+        ),
+    )