diff --git a/blunderboard-py/blunderboard.py b/blunderboard-py/blunderboard.py
new file mode 100644
index 0000000..7d70f09
--- /dev/null
+++ b/blunderboard-py/blunderboard.py
@@ -0,0 +1,84 @@
+from stockfish import Stockfish
+from playsound import playsound
+from pathlib import Path
+import random
+
+settings = {
+    "Debug Log File": "",
+    "Contempt": 0,
+    "Min Split Depth": 0,
+    "Threads": 1,
+    # More threads will make the engine stronger, but should be kept at less than the number of logical processors on
+    # your computer.
+    "Ponder": "false",
+    "Hash": 16,
+    # Default size is 16 MB. It's recommended that you increase this value, but keep it as some power of 2. E.g.,
+    # if you're fine using 2 GB of RAM, set Hash to 2048 (11th power of 2).
+    "MultiPV": 1,
+    "Skill Level": 20,
+    "Move Overhead": 10,
+    "Minimum Thinking Time": 20,
+    "Slow Mover": 100,
+    "UCI_Chess960": "false",
+    "UCI_LimitStrength": "false",
+    "UCI_Elo": 1350
+}
+
+sound_path = [Path("sounds/")]
+
+
+class BoardReader:
+    """
+    A class containing methods to read the board state through the GPIO matrix.
+    """
+
+    def __init__(self):
+        pass
+
+    # TODO WIP
+
+    def get_latest_move(self):
+        return ""
+
+
+class Game:
+    """
+    A class representing the game state
+    """
+
+    def __init__(self, settings: dict):
+        self.engine = Stockfish("/usr/local/bin/stockfish")
+        self.settings = settings
+        self.engine.update_engine_parameters(self.settings)
+        self.matrix = BoardReader()
+        self.current_evaluation = self.engine.get_evaluation()
+        self.evaluations = []
+        self.engine.set_position()
+
+        def make_move(move) -> None:
+            """
+            Makes a move on the board and updates the game state
+            :param move:
+            :return: None
+            """
+            if self.engine.is_move_correct(move):
+                self.engine.make_moves_from_current_position([move])
+                self.current_evaluation = self.engine.get_evaluation()
+                self.evaluations.append(self.current_evaluation)
+            if move_was_blunder():
+                # If the played move was a blunder play a random sound from the sound path
+                playsound(random.choice(sound_path))
+
+        def move_was_blunder() -> bool:
+            """
+            Returns true if the last move was a blunder
+            :return: bool
+            """
+            previous_evaluation = self.evaluations[-1]
+            return self.current_evaluation["value"] < (previous_evaluation["value"] + 3)  # TODO This is not a
+            # particularly good way to identify a blunder
+
+    def __str__(self):
+        return ""
+
+
diff --git a/blunderboard-py/requirements.txt b/blunderboard-py/requirements.txt
new file mode 100644
index 0000000..51ee0ef
--- /dev/null
+++ b/blunderboard-py/requirements.txt
@@ -0,0 +1,2 @@
+stockfish
+playsound
\ No newline at end of file
diff --git a/pyvenv.cfg b/pyvenv.cfg
new file mode 100644
index 0000000..8810f2d
--- /dev/null
+++ b/pyvenv.cfg
@@ -0,0 +1,3 @@
+home = /usr/bin
+include-system-site-packages = false
+version = 3.10.8