diff --git a/loaders/implementations/scrambler.py b/loaders/implementations/scrambler.py
new file mode 100644
index 0000000..9e89f6d
--- /dev/null
+++ b/loaders/implementations/scrambler.py
@@ -0,0 +1,53 @@
+import re
+import logging
+
+from requests import Request
+from radio.utils.normalize import split_artist_title
+
+logger = logging.getLogger(__name__)
+
+
+stream = True
+
+
+def form_request():
+    url = "http://stream.radio.co/sa359d3b4a/listen"
+    headers = {"Icy-MetaData": "1"}
+
+    return Request("GET", url, headers=headers)
+
+
+def parse_response(response):
+    # Find chunk size, metadata is included after each chunk
+    offset = int(response.headers.get('icy-metaint'))
+
+    # Check sane offset value (usually 16k)
+    if not (1 < offset < 64 * 1024):
+        logger.error("invalid icy-metaint value: {}")
+        return
+
+    # Skip first chunk
+    response.raw.read(offset)
+
+    # Determine length of meta data from first byte
+    length = response.raw.read(1)[0] * 16
+
+    # Check sane length value (usually 32)
+    if not (1 < length < 128):
+        logger.error("invalid meta length: {}".format(length))
+        return
+
+    # Read and decode metadata
+    meta = response.raw.read(length).decode("utf-8")
+    match = re.search("StreamTitle='(.+)';", meta)
+    if not match:
+        logger.error("metadata not found in: '{}'")
+        return
+
+    artist, title = split_artist_title(match.group(1))
+
+    # Skip commercials
+    if artist.lower() == 'radio scrambler ducati':
+        return None
+
+    return artist.title(), title