diff --git a/src/ocrmypdf/_exec/ghostscript.py b/src/ocrmypdf/_exec/ghostscript.py
index 5c357f1b..f459763a 100644
--- a/src/ocrmypdf/_exec/ghostscript.py
+++ b/src/ocrmypdf/_exec/ghostscript.py
@@ -25,28 +25,7 @@ from ocrmypdf.subprocess import get_version, run, run_polling_stderr
 
 log = logging.getLogger(__name__)
 
-missing_gs_error = """
----------------------------------------------------------------------
-This error normally occurs when ocrmypdf find can't Ghostscript.
-Please ensure Ghostscript is installed and its location is added to
-the system PATH environment variable.
-
-For details see:
-    https://ocrmypdf.readthedocs.io/en/latest/installation.html
----------------------------------------------------------------------
-"""
-
-_gswin = None
-if os.name == 'nt':
-    _gswin = which('gswin64c')
-    if not _gswin:
-        _gswin = which('gswin32c')
-        if not _gswin:
-            raise MissingDependencyError(missing_gs_error)
-    _gswin = Path(_gswin).stem
-
-GS = _gswin if _gswin else 'gs'
-del _gswin
+GS = '@gs@'
 
 
 def version():
diff --git a/src/ocrmypdf/_exec/jbig2enc.py b/src/ocrmypdf/_exec/jbig2enc.py
index 2e8a058b..65a09088 100644
--- a/src/ocrmypdf/_exec/jbig2enc.py
+++ b/src/ocrmypdf/_exec/jbig2enc.py
@@ -14,7 +14,7 @@ from ocrmypdf.subprocess import get_version, run
 
 
 def version():
-    return get_version('jbig2', regex=r'jbig2enc (\d+(\.\d+)*).*')
+    return get_version('@jbig2@', regex=r'jbig2enc (\d+(\.\d+)*).*')
 
 
 def available():
@@ -27,7 +27,7 @@ def available():
 
 def convert_group(*, cwd, infiles, out_prefix):
     args = [
-        'jbig2',
+        '@jbig2@',
         '-b',
         out_prefix,
         '-s',  # symbol mode (lossy)
@@ -46,7 +46,7 @@ def convert_group_mp(args):
 
 
 def convert_single(*, cwd, infile, outfile):
-    args = ['jbig2', '-p', infile]
+    args = ['@jbig2@', '-p', infile]
     with open(outfile, 'wb') as fstdout:
         proc = run(args, cwd=cwd, stdout=fstdout, stderr=PIPE)
     proc.check_returncode()
diff --git a/src/ocrmypdf/_exec/pngquant.py b/src/ocrmypdf/_exec/pngquant.py
index ca8a4542..d0544174 100644
--- a/src/ocrmypdf/_exec/pngquant.py
+++ b/src/ocrmypdf/_exec/pngquant.py
@@ -19,7 +19,7 @@ from ocrmypdf.subprocess import get_version, run
 
 
 def version():
-    return get_version('pngquant', regex=r'(\d+(\.\d+)*).*')
+    return get_version('@pngquant@', regex=r'(\d+(\.\d+)*).*')
 
 
 def available():
@@ -46,7 +46,7 @@ def input_as_png(input_file: Path):
 def quantize(input_file: Path, output_file: Path, quality_min: int, quality_max: int):
     with input_as_png(input_file) as input_stream:
         args = [
-            'pngquant',
+            '@pngquant@',
             '--force',
             '--skip-if-larger',
             '--quality',
diff --git a/src/ocrmypdf/_exec/tesseract.py b/src/ocrmypdf/_exec/tesseract.py
index 33ead41e..5840f7c1 100644
--- a/src/ocrmypdf/_exec/tesseract.py
+++ b/src/ocrmypdf/_exec/tesseract.py
@@ -78,7 +78,7 @@ class TesseractVersion(StrictVersion):
 
 
 def version():
-    return get_version('tesseract', regex=r'tesseract\s(.+)')
+    return get_version('@tesseract@', regex=r'tesseract\s(.+)')
 
 
 def has_user_words():
@@ -100,7 +100,7 @@ def get_languages():
         msg += output
         return msg
 
-    args_tess = ['tesseract', '--list-langs']
+    args_tess = ['@tesseract@', '--list-langs']
     try:
         proc = run(
             args_tess,
@@ -122,7 +122,7 @@ def get_languages():
 
 
 def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
-    args = ['tesseract']
+    args = ['@tesseract@']
     if langs:
         args.extend(['-l', '+'.join(langs)])
     if engine_mode is not None:
diff --git a/src/ocrmypdf/_exec/unpaper.py b/src/ocrmypdf/_exec/unpaper.py
index 3c3ae72c..d269966a 100644
--- a/src/ocrmypdf/_exec/unpaper.py
+++ b/src/ocrmypdf/_exec/unpaper.py
@@ -31,7 +31,7 @@ log = logging.getLogger(__name__)
 
 
 def version() -> str:
-    return get_version('unpaper')
+    return get_version('@unpaper@')
 
 
 def _setup_unpaper_io(tmpdir: Path, input_file: Path) -> Tuple[Path, Path]:
@@ -71,7 +71,7 @@ def _setup_unpaper_io(tmpdir: Path, input_file: Path) -> Tuple[Path, Path]:
 def run(
     input_file: Path, output_file: Path, *, dpi: DecFloat, mode_args: List[str]
 ) -> None:
-    args_unpaper = ['unpaper', '-v', '--dpi', str(round(dpi, 6))] + mode_args
+    args_unpaper = ['@unpaper@', '-v', '--dpi', str(round(dpi, 6))] + mode_args
 
     with TemporaryDirectory() as tmpdir:
         input_pnm, output_pnm = _setup_unpaper_io(Path(tmpdir), input_file)
diff --git a/src/ocrmypdf/leptonica.py b/src/ocrmypdf/leptonica.py
index e4814f1a..fdaf7ea4 100644
--- a/src/ocrmypdf/leptonica.py
+++ b/src/ocrmypdf/leptonica.py
@@ -33,14 +33,7 @@ from ocrmypdf.lib._leptonica import ffi
 
 logger = logging.getLogger(__name__)
 
-if os.name == 'nt':
-    from ocrmypdf.subprocess._windows import shim_env_path
-
-    libname = 'liblept-5'
-    os.environ['PATH'] = shim_env_path()
-else:
-    libname = 'lept'
-_libpath = find_library(libname)
+_libpath = '@liblept@'
 if not _libpath:
     raise MissingDependencyError(
         """