diff --git a/backend/tests/test_deepagents_e2e.py b/backend/tests/test_deepagents_e2e.py
new file mode 100644
index 0000000..fbb7447
--- /dev/null
+++ b/backend/tests/test_deepagents_e2e.py
@@ -0,0 +1,79 @@
+import pytest
+from unittest.mock import MagicMock
+from ea_chatbot.graph.workflow import create_workflow
+from ea_chatbot.graph.state import AgentState
+from langchain_core.messages import AIMessage, HumanMessage
+
+def test_deepagents_multi_worker_sequential_flow():
+    """Verify that the Orchestrator can handle a sequence of different workers."""
+    
+    mock_analyzer = MagicMock()
+    mock_planner = MagicMock()
+    mock_delegate = MagicMock()
+    mock_analyst = MagicMock()
+    mock_researcher = MagicMock()
+    mock_reflector = MagicMock()
+    mock_synthesizer = MagicMock()
+    
+    # 1. Analyzer: Plan
+    mock_analyzer.return_value = {"next_action": "plan"}
+    
+    # 2. Planner: Two tasks
+    mock_planner.return_value = {
+        "checklist": [
+            {"task": "Get Numbers", "worker": "data_analyst"},
+            {"task": "Get Facts", "worker": "researcher"}
+        ],
+        "current_step": 0
+    }
+    
+    # 3. Delegate & Reflector: Loop through tasks
+    mock_delegate.side_effect = [
+        {"next_action": "data_analyst"}, # Step 0
+        {"next_action": "researcher"},   # Step 1
+        {"next_action": "summarize"}     # Step 2 (done)
+    ]
+    
+    mock_analyst.return_value = {"messages": [AIMessage(content="Analyst summary")], "vfs": {"data.csv": "..."}}
+    mock_researcher.return_value = {"messages": [AIMessage(content="Researcher summary")]}
+    
+    mock_reflector.side_effect = [
+        {"current_step": 1, "next_action": "delegate"}, # Done with Analyst
+        {"current_step": 2, "next_action": "delegate"}  # Done with Researcher
+    ]
+    
+    mock_synthesizer.return_value = {
+        "messages": [AIMessage(content="Final multi-agent response")],
+        "next_action": "end"
+    }
+    
+    app = create_workflow(
+        query_analyzer=mock_analyzer,
+        planner=mock_planner,
+        delegate=mock_delegate,
+        data_analyst_worker=mock_analyst,
+        researcher_worker=mock_researcher,
+        reflector=mock_reflector,
+        synthesizer=mock_synthesizer
+    )
+    
+    initial_state = AgentState(
+        messages=[HumanMessage(content="Numbers and facts please")],
+        question="Numbers and facts please",
+        analysis={},
+        next_action="",
+        iterations=0,
+        checklist=[],
+        current_step=0,
+        vfs={},
+        plots=[],
+        dfs={}
+    )
+    
+    final_state = app.invoke(initial_state, config={"recursion_limit": 30})
+    
+    assert mock_analyst.called
+    assert mock_researcher.called
+    assert mock_reflector.call_count == 2
+    assert "Final multi-agent response" in [m.content for m in final_state["messages"]]
+    assert final_state["current_step"] == 2