diff --git a/contrib/weights-from-festa-sommariva.py b/contrib/weights-from-festa-sommariva.py
index baab0f9e..ca720466 100644
--- a/contrib/weights-from-festa-sommariva.py
+++ b/contrib/weights-from-festa-sommariva.py
@@ -76,17 +76,17 @@ def generate_festa_sommariva_quadrature_rules(outfile):
     #
     #   DEGREE:  <degree>
 
-    _re_degree = re.compile(r"DEGREE:\s+(\d+)")
-    _re_xw = re.compile(r"xw\s?=\s?\[(.+?)\];", re.DOTALL)
+    re_degree = re.compile(r"DEGREE:\s+(\d+)")
+    re_xw = re.compile(r"xw\s?=\s?\[(.+?)\];", re.DOTALL)
 
     # NOTE: some degrees have multiple quadrature rules with a repeated
     # header, so we just take the unique ones here
     degrees = np.unique(
-            np.fromiter(_re_degree.findall(mfile), dtype=np.int64)
+            np.fromiter(re_degree.findall(mfile), dtype=np.int64)
             )
 
     rules = {}
-    for imatch, match in enumerate(_re_xw.findall(mfile)):
+    for imatch, match in enumerate(re_xw.findall(mfile)):
         d = degrees[imatch]
         assert d == imatch + 1, (d, imatch + 1)
 
diff --git a/modepy/modal_decay.py b/modepy/modal_decay.py
index d5fe13a6..bbbb3a7e 100644
--- a/modepy/modal_decay.py
+++ b/modepy/modal_decay.py
@@ -103,7 +103,9 @@ def make_mode_number_vector(
     return mode_number_vector
 
 
-def create_decay_baseline(mode_number_vector: np.ndarray, n: int) -> np.ndarray:
+def create_decay_baseline(
+        mode_number_vector: np.ndarray[tuple[int, ...], np.dtype[np.floating]],
+        n: int) -> np.ndarray[tuple[int, ...], np.dtype[np.floating]]:
     """Create a vector of modal coefficients that exhibit 'optimal'
     (:math:`k^{-N}`) decay.
     """
@@ -111,10 +113,8 @@ def create_decay_baseline(mode_number_vector: np.ndarray, n: int) -> np.ndarray:
     zeros = mode_number_vector == 0
 
     modal_coefficients = mode_number_vector**(-n)
-    modal_coefficients[zeros] = 1  # irrelevant, just keeps log from NaNing
-
-    # NOTE: mypy seems to be confused by the __itruediv__ argument types
-    modal_coefficients /= la.norm(modal_coefficients)   # type: ignore[arg-type,misc]
+    modal_coefficients[zeros] = 1.0  # irrelevant, just keeps log from NaNing
+    modal_coefficients /= la.norm(modal_coefficients)
 
     return modal_coefficients
 
diff --git a/modepy/quadrature/clenshaw_curtis.py b/modepy/quadrature/clenshaw_curtis.py
index 80bdd118..35bf448b 100644
--- a/modepy/quadrature/clenshaw_curtis.py
+++ b/modepy/quadrature/clenshaw_curtis.py
@@ -49,7 +49,7 @@ def _make_clenshaw_curtis_nodes_and_weights(n: int) -> tuple[np.ndarray, np.ndar
     # Clenshaw-Curtis weights
     w = np.concatenate([2 / N / (N - 2), 1 / N[-1:], np.zeros(m)])
     w = 0 - w[:-1] - w[-1:0:-1]
-    g0 = -np.ones(n)
+    g0: np.ndarray[tuple[int, ...], np.dtype[np.floating]] = -np.ones(n)
     g0[r] = g0[r] + n
     g0[m] = g0[m] + n
     g0 = g0 / (n**2 - 1 + (n % 2))
diff --git a/modepy/quadrature/jacobi_gauss.py b/modepy/quadrature/jacobi_gauss.py
index 8fd41597..dc56eb28 100644
--- a/modepy/quadrature/jacobi_gauss.py
+++ b/modepy/quadrature/jacobi_gauss.py
@@ -50,7 +50,6 @@
 THE SOFTWARE.
 """
 
-
 import numpy as np
 import numpy.linalg as la
 
@@ -267,7 +266,7 @@ def jacobi_gauss_lobatto_nodes(
     Exact to degree :math:`2N - 3`.
     """
 
-    x = np.zeros((N + 1,))
+    x: np.ndarray[tuple[int, ...], np.dtype[np.floating]] = np.zeros((N + 1,))
     if N == 0:
         x[0] = 0
         return x