diff --git a/src/120 bpm amen break.mp3 b/src/120 bpm amen break.mp3
new file mode 100644
index 0000000..035de02
Binary files /dev/null and b/src/120 bpm amen break.mp3 differ
diff --git a/src/assets/style.tcss b/src/assets/style.tcss
new file mode 100644
index 0000000..e0cdb13
--- /dev/null
+++ b/src/assets/style.tcss
@@ -0,0 +1,3 @@
+App {
+    layers: bottom top;
+}
\ No newline at end of file
diff --git a/src/main.py b/src/main.py
index 54b26f6..6c32ff6 100644
--- a/src/main.py
+++ b/src/main.py
@@ -1,19 +1,12 @@
 from ui.app import AppUI
-from project import Project, ProjectChannel, ChannelChunk, AudioChannelChunk
-from ui.widgets.chunk_types.audio import AudioChunk
-import librosa
+from project import Project
 
 
 if __name__ == "__main__":
-    test_project = Project(
-        [
-            ProjectChannel("my channel", chunks=[
-                ChannelChunk(name="hi"),
-                AudioChannelChunk(librosa.load("cool sample.mp3", sr=None, mono=False))
-            ])
-        ]
-    )
+    print("Loading project...")
+    test_project = Project.from_file("test_project.tdp")
     
     # start the ui
+    print("Starting UI...")
     app = AppUI(test_project)
     app.run()
\ No newline at end of file
diff --git a/src/project.py b/src/project.py
index a00dd50..c8d7440 100644
--- a/src/project.py
+++ b/src/project.py
@@ -38,15 +38,17 @@ class ChannelChunk:
         }
 
 class AudioChannelChunk(ChannelChunk):
-    def __init__(self, audio_data: np.ndarray, position: float = 0.0, name: str = "Sample"):
+    def __init__(self, audio_data: np.ndarray, sample_rate: int, position: float = 0.0, name: str = "Sample"):
         super().__init__(position, name, chunk_type=ChunkType.AUDIO)
         self.audio_data = audio_data
+        self.sample_rate = sample_rate
         
     def from_json(json: dict) -> ChannelChunk:
         return AudioChannelChunk(
             name = json["name"],
             position = json["position"],
-            audio_data = json["audio_data"]
+            audio_data = json["audio_data"],
+            sample_rate = json["sample_rate"]
         )
         
     def to_json(self):
@@ -54,7 +56,8 @@ class AudioChannelChunk(ChannelChunk):
             "type": self.chunk_type.value,
             "name": self.name,
             "position": self.position,
-            "audio_data": self.audio_data
+            "audio_data": self.audio_data,
+            "sample_rate": self.sample_rate
         }
 
 chunk_type_associations = {
diff --git a/src/test_project.tdp b/src/test_project.tdp
new file mode 100644
index 0000000..2cfefce
Binary files /dev/null and b/src/test_project.tdp differ
diff --git a/src/ui/app.py b/src/ui/app.py
index ee6ce7f..4020b68 100644
--- a/src/ui/app.py
+++ b/src/ui/app.py
@@ -7,6 +7,8 @@ from ui.widgets.project_settings import ProjectSettings
 
 
 class AppUI(App):    
+    CSS_PATH = "../assets/style.tcss"
+    
     def __init__(self, project):
         super().__init__()
         self.zoom_level = 0.05
diff --git a/src/ui/widgets/chunk_types/audio.py b/src/ui/widgets/chunk_types/audio.py
index ac65715..732eba3 100644
--- a/src/ui/widgets/chunk_types/audio.py
+++ b/src/ui/widgets/chunk_types/audio.py
@@ -25,11 +25,11 @@ class AudioChunk(Chunk):
     }
     """
 
-    def __init__(self, file_path: str, chunk_name: str = "Sample"):
+    def __init__(self, audio_data: np.ndarray, sample_rate: int, chunk_name: str = "Sample"):
         super().__init__(chunk_name)
         
-        self.file_path = file_path
-        self.audio, self.sample_rate = librosa.load(self.file_path, sr=None, mono=False)
+        self.audio = audio_data
+        self.sample_rate = sample_rate
         
         self.num_channels = None
         if len(self.audio.shape) == 1:
@@ -42,7 +42,7 @@ class AudioChunk(Chunk):
         self.meter = pyln.Meter(self.sample_rate)
         self.loudness_values = []
         
-        self.styles.width = (self.num_samples / self.sample_rate) / self.app.zoom_level
+        self.styles.width = int((self.num_samples / self.sample_rate) / self.app.zoom_level)
     
     def on_mount(self):
         for plot in self.query(PlotWidget):
diff --git a/src/ui/widgets/play_head.py b/src/ui/widgets/play_head.py
new file mode 100644
index 0000000..769bba6
--- /dev/null
+++ b/src/ui/widgets/play_head.py
@@ -0,0 +1,14 @@
+from textual.widgets import Rule
+
+
+class PlayHead(Rule):
+    DEFAULT_CSS = """
+    PlayHead.-vertical {
+        layer: top;
+        margin: 0;
+        color: $accent;
+    }
+    """
+    
+    def __init__(self):
+        super().__init__(orientation="vertical")
\ No newline at end of file
diff --git a/src/ui/widgets/sidebar.py b/src/ui/widgets/sidebar.py
index a74974a..00c704e 100644
--- a/src/ui/widgets/sidebar.py
+++ b/src/ui/widgets/sidebar.py
@@ -23,6 +23,12 @@ class Sidebar(Vertical):
     
     def compose(self) -> ComposeResult:
         with VerticalScroll(id="channels"):
-            yield Channel()
-            yield Channel()
+            for channel in self.app.project.channels:
+                yield Channel(
+                    channel.name,
+                    channel.mute,
+                    channel.solo,
+                    channel.pan,
+                    channel.volume
+                )
             yield Button("+ New Channel", variant="success", id="add-channel")
\ No newline at end of file
diff --git a/src/ui/widgets/timeline.py b/src/ui/widgets/timeline.py
index 17895dc..fca9f6e 100644
--- a/src/ui/widgets/timeline.py
+++ b/src/ui/widgets/timeline.py
@@ -1,29 +1,84 @@
-from textual.containers import Vertical,  VerticalScroll, Horizontal, VerticalGroup
-from textual.widgets import Sparkline
+from textual.containers import Vertical, VerticalScroll, Horizontal, VerticalGroup, HorizontalGroup
+from textual.widgets import Rule
 from textual.app import ComposeResult
 
-from ui.widgets.chunk_types.audio import AudioChunk
+from ui.widgets.chunk_types.audio import AudioChunk, Chunk
+from ui.widgets.play_head import PlayHead
+from project import ChunkType
 
 
 class TimelineRow(Horizontal):
     DEFAULT_CSS = """
     TimelineRow {
         background: $surface-lighten-1;
-        height: 8;
+        max-height: 8;
         margin-bottom: 1;
     }
     """
 
-class Timeline(VerticalScroll):
+class Timeline(Vertical):
     DEFAULT_CSS = """
     Timeline {
-        padding: 1 0;
-        hatch: "-" $surface-lighten-1;
+        
+        #rows {
+            hatch: "-" $surface-lighten-1;
+            padding: 1 0;
+            
+            .beat-line {
+                color: $surface-lighten-1;
+                
+            }
+            
+            .bar-line {
+                color: $surface-lighten-2;
+                
+            }
+            
+            .beat-line, .bar-line {
+                dock: left;
+                margin: 0;
+            }
+            
+            
+        }
+        
+        PlayHead {
+            layer: top;
+        }
     }
     """
     
+    def __init__(self):
+        super().__init__()
+        
+        self.bar_offset = self.app.project.bpm / 8 * (0.0333 / self.app.zoom_level)
+    
     def compose(self) -> ComposeResult:
-        with TimelineRow():
-            yield AudioChunk("cool sample.mp3")
-        with TimelineRow():
-            yield AudioChunk("cool sample 2.mp3")
\ No newline at end of file
+        
+        
+        with VerticalScroll(id="rows"):
+            
+            for i in range(1, 17):
+                bar = None
+                if (i) % 4 == 0:
+                    bar = Rule.vertical(classes="bar-line", line_style="double")
+                else:
+                    bar = Rule.vertical(classes="beat-line")
+                    
+                bar.offset = (self.bar_offset * i, 0)
+                    
+                yield bar
+            
+            for channel in self.app.project.channels:
+                
+                
+                with TimelineRow():
+                    for chunk in channel.chunks:
+                        if chunk.chunk_type == ChunkType.CHUNK:
+                            yield Chunk(chunk_name=chunk.name)
+                        elif chunk.chunk_type == ChunkType.AUDIO:
+                            yield AudioChunk(chunk.audio_data, chunk.sample_rate, chunk.name)
+             
+            
+             
+        #yield PlayHead()
\ No newline at end of file