diff --git a/tests/test_engine/README.md b/tests/test_engine/README.md
new file mode 100644
index 0000000..a293176
--- /dev/null
+++ b/tests/test_engine/README.md
@@ -0,0 +1,195 @@
+# Engine Module Test Suite
+
+This directory contains comprehensive tests for the vector search engine module, covering all components from the abstract base engine to concrete implementations and the factory pattern.
+
+## Test Structure
+
+### Core Test Files
+
+- **`test_base_engine.py`** - Tests for the abstract `BaseEngine` class
+- **`test_qdrant_engine.py`** - Tests for the `QdrantEngine` implementation
+- **`test_factory.py`** - Tests for the engine factory and `EngineType` enum
+- **`test_integration.py`** - End-to-end integration tests
+- **`conftest.py`** - Shared fixtures and test configuration
+
+## Test Coverage
+
+### BaseEngine Tests (`test_base_engine.py`)
+- ✅ Abstract class enforcement - ensures BaseEngine cannot be instantiated
+- ✅ Abstract method verification - validates required method signatures
+- ✅ Generic typing constraints - tests TypeVar functionality
+- ✅ Semantic search workflow - tests the complete search flow
+- ✅ Condition transformation - tests with various condition types
+- ✅ Response transformation - tests data structure conversion
+- ✅ Parameter validation - tests default values and edge cases
+
+### QdrantEngine Tests (`test_qdrant_engine.py`)
+- ✅ Inheritance verification - ensures proper BaseEngine inheritance
+- ✅ Generic type parameters - validates `BaseEngine[list[ScoredPoint], Filter]`
+- ✅ Condition transformation - tests conversion to Qdrant Filter objects
+  - Match conditions → `MatchValue`
+  - MatchAny conditions → `MatchAny`
+  - MatchExclude conditions → `MatchExcept`
+- ✅ Response transformation - tests `ScoredPoint` to `SearchRow` conversion
+- ✅ Null payload filtering - ensures entries with null payloads are excluded
+- ✅ Client interaction - mocks and verifies Qdrant client calls
+- ✅ Error propagation - ensures client exceptions bubble up correctly
+- ✅ Initialization - tests Settings and AsyncQdrantClient setup
+
+### Factory Tests (`test_factory.py`)
+- ✅ Backend type enumeration - tests `Backend` enum values and behavior
+- ✅ Factory function typing - validates overload signatures for type safety
+- ✅ Engine instantiation - tests creation of concrete engine instances
+- ✅ Error handling - validates behavior with invalid inputs
+- ✅ Caching behavior - ensures `@cache` decorator works correctly (same instances)
+- ✅ COSMOS engine handling - tests NotImplementedError for unimplemented engines
+- ✅ String enum behavior - tests StrEnum functionality and JSON serialization
+
+### Integration Tests (`test_integration.py`)
+- ✅ Complete workflow - factory → conditions → search → response transformation
+- ✅ Parameter passing - verifies correct parameter flow through all layers
+- ✅ Complex conditions - tests multiple condition types together
+- ✅ Large result sets - tests handling of 100+ search results
+- ✅ Edge cases - empty conditions, null payloads, error scenarios
+- ✅ Named vectors - tests support for Qdrant NamedVector objects
+- ✅ Multiple engine instances - tests independence and concurrent usage
+
+## Test Fixtures (`conftest.py`)
+
+### Automatic Fixtures
+- `clear_engine_cache` - Auto-clears `@cache` decorator before/after each test
+
+### Mock Objects
+- `mock_qdrant_client` - AsyncMock for Qdrant client with default responses
+- `mock_settings` - Mock Settings object with test configuration
+- `mock_qdrant_engine_dependencies` - Complete mocked environment
+
+### Sample Data
+- `sample_embedding` - Standard test embedding vector
+- `sample_conditions` - Common condition objects for testing
+- `sample_scored_points` - Realistic ScoredPoint objects
+- `sample_search_rows` - Expected SearchRow outputs
+- `qdrant_filter_single/multiple` - Pre-built Qdrant Filter objects
+
+## Running Tests
+
+### Run All Engine Tests
+```bash
+uv run pytest tests/test_engine/ -v
+```
+
+### Run Specific Test File
+```bash
+uv run pytest tests/test_engine/test_base_engine.py -v
+uv run pytest tests/test_engine/test_qdrant_engine.py -v
+uv run pytest tests/test_engine/test_factory.py -v
+uv run pytest tests/test_engine/test_integration.py -v
+```
+
+### Run Specific Test Class or Method
+```bash
+uv run pytest tests/test_engine/test_factory.py::TestEngineFactory::test_get_engine_qdrant -v
+uv run pytest tests/test_engine/test_integration.py::TestEngineIntegration -v
+```
+
+### Coverage Report
+```bash
+uv run pytest tests/test_engine/ --cov=src/vector_search_mcp/engine --cov-report=html
+```
+
+## Test Patterns Used
+
+### Mocking Strategy
+- **External dependencies** - All external services (Qdrant client, Settings) are mocked
+- **Dependency injection** - Tests inject mocks through constructor parameters
+- **Return value control** - Mocks return predictable test data for assertions
+
+### Async Testing
+- Uses `@pytest.mark.asyncio` for async method testing
+- `AsyncMock` objects for async client methods
+- Proper await/async syntax throughout
+
+### Type Testing
+- Generic type parameter validation
+- Overload signature verification  
+- Runtime type checking where applicable
+
+### Error Testing
+- Exception propagation validation
+- Invalid input handling
+- Boundary condition testing
+
+## Key Test Insights
+
+### Generic Typing Validation
+The tests verify that the generic `BaseEngine[ResponseType, ConditionType]` pattern works correctly:
+- `QdrantEngine` is typed as `BaseEngine[list[ScoredPoint], Filter]`
+- Type checkers can verify correct usage at compile time
+- Runtime behavior matches type declarations
+
+### Factory Pattern Testing
+The overload tests ensure proper type inference:
+```python
+# Type checker knows this returns QdrantEngine
+engine = get_engine(Backend.QDRANT)
+
+# Type checker knows this returns BaseEngine (generic)
+backend: Backend = some_variable
+engine = get_engine(backend)
+```
+
+### Caching Behavior
+Tests verify that the `@cache` decorator works correctly:
+```python
+# Both calls return the same instance
+engine1 = get_engine(Backend.QDRANT)
+engine2 = get_engine(Backend.QDRANT)
+assert engine1 is engine2  # Same instance due to caching
+```
+
+### Integration Flow Validation
+Integration tests verify the complete data flow:
+1. `get_engine()` creates proper engine instance
+2. `semantic_search()` calls `transform_conditions()`
+3. Transformed conditions passed to `run_similarity_query()`
+4. Query response processed by `transform_response()`
+5. Final `SearchRow` objects returned to caller
+
+## Maintenance Notes
+
+### Adding New Engine Types
+When adding new engines:
+1. Add enum value to `Backend`
+2. Add overload signature to `get_engine()`
+3. Update factory tests for new count and behavior
+4. Create engine-specific test file following `test_qdrant_engine.py` pattern
+5. Remember that `@cache` decorator will cache instances per backend type
+
+### Mock Updates
+If engine interfaces change:
+1. Update fixture return types in `conftest.py`
+2. Verify mock method signatures match real implementations
+3. Update integration tests for new parameter flows
+4. Ensure cache clearing fixture handles any new caching behavior
+
+### Performance Testing
+Current tests focus on correctness. For performance testing:
+- Use `pytest-benchmark` for timing critical paths
+- Test with realistic data sizes (1000+ embeddings)
+- Mock network I/O but measure transformation logic
+- Consider cache warming effects when benchmarking
+
+## Recent Fixes Applied
+
+### After Formatter Changes
+The following issues were resolved after code formatting:
+
+1. **Enum Rename**: `EngineType` → `Backend` - All tests updated
+2. **Caching Addition**: `@cache` decorator added to `get_engine()` 
+   - Tests updated to expect same instances (not different ones)
+   - Auto-cache clearing fixture added to `conftest.py`
+3. **Mock Isolation**: Improved mock setup to prevent real network calls
+   - Proper patch contexts in all integration tests
+   - Cache clearing ensures clean test state
+
+All 62 tests now pass successfully! 🎉
\ No newline at end of file
diff --git a/tests/test_engine/__init__.py b/tests/test_engine/__init__.py
new file mode 100644
index 0000000..ae31e71
--- /dev/null
+++ b/tests/test_engine/__init__.py
@@ -0,0 +1 @@
+# Test package for engine module
diff --git a/tests/test_engine/conftest.py b/tests/test_engine/conftest.py
new file mode 100644
index 0000000..da6432c
--- /dev/null
+++ b/tests/test_engine/conftest.py
@@ -0,0 +1,163 @@
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+from qdrant_client import models
+
+from vector_search_mcp.engine import get_engine
+from vector_search_mcp.models import Match, MatchAny, MatchExclude, SearchRow
+
+
+@pytest.fixture(autouse=True)
+def clear_engine_cache():
+    """Clear the engine cache before each test for proper isolation"""
+    get_engine.cache_clear()
+    yield
+    get_engine.cache_clear()
+
+
+@pytest.fixture
+def mock_qdrant_client():
+    """Create a mock Qdrant client for testing"""
+    client = AsyncMock()
+
+    # Default search response
+    client.search.return_value = [
+        models.ScoredPoint(
+            id=1,
+            score=0.95,
+            payload={"text": "Test document 1", "category": "test"},
+            version=1,
+        ),
+        models.ScoredPoint(
+            id=2,
+            score=0.85,
+            payload={"text": "Test document 2", "category": "test"},
+            version=1,
+        ),
+    ]
+
+    return client
+
+
+@pytest.fixture
+def mock_settings():
+    """Create mock settings for testing"""
+    settings = MagicMock()
+    settings.url = "http://localhost:6333"
+    settings.api_key = "test_api_key"
+    return settings
+
+
+@pytest.fixture
+def sample_embedding():
+    """Provide a sample embedding vector for testing"""
+    return [0.1, 0.2, 0.3, 0.4, 0.5]
+
+
+@pytest.fixture
+def sample_conditions():
+    """Provide sample conditions for testing"""
+    return [
+        Match(key="category", value="document"),
+        MatchAny(key="tags", any=["python", "rust"]),
+        MatchExclude(key="status", exclude=["deleted"]),
+    ]
+
+
+@pytest.fixture
+def sample_scored_points():
+    """Provide sample ScoredPoint objects for testing"""
+    return [
+        models.ScoredPoint(
+            id=1,
+            score=0.95,
+            payload={"text": "First document", "category": "tech"},
+            version=1,
+        ),
+        models.ScoredPoint(
+            id=2,
+            score=0.87,
+            payload={"text": "Second document", "category": "science"},
+            version=1,
+        ),
+        models.ScoredPoint(
+            id=3,
+            score=0.75,
+            payload=None,  # This should be filtered out
+            version=1,
+        ),
+    ]
+
+
+@pytest.fixture
+def sample_search_rows():
+    """Provide sample SearchRow objects for testing"""
+    return [
+        SearchRow(
+            chunk_id="1",
+            score=0.95,
+            payload={"text": "First document", "category": "tech"},
+        ),
+        SearchRow(
+            chunk_id="2",
+            score=0.87,
+            payload={"text": "Second document", "category": "science"},
+        ),
+    ]
+
+
+@pytest.fixture
+def mock_qdrant_engine_dependencies():
+    """Mock all external dependencies for QdrantEngine"""
+    with (
+        patch("vector_search_mcp.engine.qdrant_engine.Settings") as mock_settings_class,
+        patch(
+            "vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"
+        ) as mock_client_class,
+    ):
+        # Setup mock settings
+        mock_settings = MagicMock()
+        mock_settings.url = "http://localhost:6333"
+        mock_settings.api_key = "test_api_key"
+        mock_settings_class.return_value = mock_settings
+
+        # Setup mock client
+        mock_client = AsyncMock()
+        mock_client_class.return_value = mock_client
+
+        yield {
+            "settings_class": mock_settings_class,
+            "client_class": mock_client_class,
+            "settings": mock_settings,
+            "client": mock_client,
+        }
+
+
+@pytest.fixture
+def qdrant_filter_single():
+    """Create a single-condition Qdrant filter for testing"""
+    return models.Filter(
+        must=[
+            models.FieldCondition(
+                key="category", match=models.MatchValue(value="document")
+            )
+        ]
+    )
+
+
+@pytest.fixture
+def qdrant_filter_multiple():
+    """Create a multi-condition Qdrant filter for testing"""
+    return models.Filter(
+        must=[
+            models.FieldCondition(
+                key="category", match=models.MatchValue(value="document")
+            ),
+            models.FieldCondition(
+                key="tags", match=models.MatchAny(any=["python", "rust"])
+            ),
+            models.FieldCondition(
+                key="status", match=models.MatchExcept(**{"except": ["deleted"]})
+            ),
+        ]
+    )
diff --git a/tests/test_engine/test_base_engine.py b/tests/test_engine/test_base_engine.py
new file mode 100644
index 0000000..87246e7
--- /dev/null
+++ b/tests/test_engine/test_base_engine.py
@@ -0,0 +1,227 @@
+from typing import Any
+
+import pytest
+
+from vector_search_mcp.engine.base_engine import BaseEngine
+from vector_search_mcp.models import Condition, Match, MatchAny, SearchRow
+
+
+class MockEngine(BaseEngine[dict[str, Any], str]):
+    """Mock engine for testing BaseEngine abstract functionality"""
+
+    def __init__(self):
+        self.transform_conditions_called = False
+        self.transform_response_called = False
+        self.run_similarity_query_called = False
+
+    def transform_conditions(self, conditions: list[Condition] | None) -> str | None:
+        self.transform_conditions_called = True
+        if not conditions:
+            return None
+        return f"transformed_{len(conditions)}_conditions"
+
+    def transform_response(self, response: dict[str, Any]) -> list[SearchRow]:
+        self.transform_response_called = True
+        return [
+            SearchRow(
+                chunk_id=str(i),
+                score=response.get(f"score_{i}", 0.5),
+                payload={"text": f"result_{i}"},
+            )
+            for i in range(response.get("count", 1))
+        ]
+
+    async def run_similarity_query(
+        self,
+        embedding: list[float],
+        collection: str,
+        limit: int = 10,
+        conditions: str | None = None,
+        threshold: float | None = None,
+    ) -> dict[str, Any]:
+        self.run_similarity_query_called = True
+        return {
+            "count": min(limit, 3),
+            "collection": collection,
+            "conditions": conditions,
+            "threshold": threshold,
+            "score_0": 0.95,
+            "score_1": 0.85,
+            "score_2": 0.75,
+        }
+
+
+class TestBaseEngine:
+    """Test suite for BaseEngine abstract class"""
+
+    def test_base_engine_is_abstract(self):
+        """Test that BaseEngine cannot be instantiated directly"""
+        with pytest.raises(TypeError):
+            BaseEngine()
+
+    def test_base_engine_abc_methods(self):
+        """Test that BaseEngine has the correct abstract methods"""
+        abstract_methods = BaseEngine.__abstractmethods__
+        expected_methods = {
+            "transform_conditions",
+            "transform_response",
+            "run_similarity_query",
+        }
+        assert abstract_methods == expected_methods
+
+    def test_mock_engine_instantiation(self):
+        """Test that mock engine can be instantiated"""
+        engine = MockEngine()
+        assert isinstance(engine, BaseEngine)
+        assert not engine.transform_conditions_called
+        assert not engine.transform_response_called
+        assert not engine.run_similarity_query_called
+
+    def test_transform_conditions_with_none(self):
+        """Test transform_conditions with None input"""
+        engine = MockEngine()
+        result = engine.transform_conditions(None)
+
+        assert result is None
+        assert engine.transform_conditions_called
+
+    def test_transform_conditions_with_conditions(self):
+        """Test transform_conditions with actual conditions"""
+        engine = MockEngine()
+        conditions = [
+            Match(key="category", value="test"),
+            MatchAny(key="tags", any=["tag1", "tag2"]),
+        ]
+
+        result = engine.transform_conditions(conditions)
+
+        assert result == "transformed_2_conditions"
+        assert engine.transform_conditions_called
+
+    def test_transform_response(self):
+        """Test transform_response with mock data"""
+        engine = MockEngine()
+        mock_response = {"count": 2, "score_0": 0.9, "score_1": 0.8}
+
+        result = engine.transform_response(mock_response)
+
+        assert len(result) == 2
+        assert all(isinstance(row, SearchRow) for row in result)
+        assert result[0].chunk_id == "0"
+        assert result[0].score == 0.9
+        assert result[1].chunk_id == "1"
+        assert result[1].score == 0.8
+        assert engine.transform_response_called
+
+    @pytest.mark.asyncio
+    async def test_run_similarity_query(self):
+        """Test run_similarity_query with various parameters"""
+        engine = MockEngine()
+        embedding = [0.1, 0.2, 0.3, 0.4, 0.5]
+
+        result = await engine.run_similarity_query(
+            embedding=embedding,
+            collection="test_collection",
+            limit=5,
+            conditions="test_conditions",
+            threshold=0.7,
+        )
+
+        assert isinstance(result, dict)
+        assert result["collection"] == "test_collection"
+        assert result["conditions"] == "test_conditions"
+        assert result["threshold"] == 0.7
+        assert result["count"] == 3  # min(limit=5, 3)
+        assert engine.run_similarity_query_called
+
+    @pytest.mark.asyncio
+    async def test_semantic_search_full_flow(self):
+        """Test the complete semantic_search flow"""
+        engine = MockEngine()
+        vector = [0.1, 0.2, 0.3, 0.4, 0.5]
+        conditions = [Match(key="status", value="active")]
+
+        result = await engine.semantic_search(
+            embedding=vector,
+            collection="test_collection",
+            limit=2,
+            conditions=conditions,
+            threshold=0.8,
+        )
+
+        # Verify all methods were called
+        assert engine.transform_conditions_called
+        assert engine.run_similarity_query_called
+        assert engine.transform_response_called
+
+        # Verify result structure
+        assert isinstance(result, list)
+        assert len(result) == 2
+        assert all(isinstance(row, SearchRow) for row in result)
+        assert result[0].score == 0.95
+        assert result[1].score == 0.85
+
+    @pytest.mark.asyncio
+    async def test_semantic_search_with_none_conditions(self):
+        """Test semantic_search with None conditions"""
+        engine = MockEngine()
+        vector = [0.1, 0.2, 0.3]
+
+        result = await engine.semantic_search(
+            embedding=vector, collection="test_collection"
+        )
+
+        assert engine.transform_conditions_called
+        assert engine.run_similarity_query_called
+        assert engine.transform_response_called
+        assert isinstance(result, list)
+        assert len(result) == 3  # default limit from mock
+
+    @pytest.mark.asyncio
+    async def test_semantic_search_default_parameters(self):
+        """Test semantic_search with default parameters"""
+        engine = MockEngine()
+        vector = [0.1, 0.2, 0.3]
+
+        result = await engine.semantic_search(
+            embedding=vector, collection="test_collection"
+        )
+
+        # Check that defaults were used (limit=10, conditions=None, threshold=None)
+        assert isinstance(result, list)
+        assert len(result) <= 10  # Should respect limit
+
+    def test_typing_constraints(self):
+        """Test that the generic typing works correctly"""
+        engine = MockEngine()
+
+        # Verify the engine has the correct generic types
+        assert hasattr(engine, "transform_conditions")
+        assert hasattr(engine, "transform_response")
+        assert hasattr(engine, "run_similarity_query")
+
+        # The mock engine should work with dict[str, Any] and str types
+        conditions_result = engine.transform_conditions([])
+        assert conditions_result is None or isinstance(conditions_result, str)
+
+        response_result = engine.transform_response({"test": "data"})
+        assert isinstance(response_result, list)
+        assert all(isinstance(item, SearchRow) for item in response_result)
+
+
+class IncompleteEngine(BaseEngine[str, int]):
+    """Incomplete engine implementation for testing abstract method enforcement"""
+
+    def transform_conditions(self, conditions: list[Condition] | None) -> int | None:
+        return None
+
+    # Missing transform_response and run_similarity_query
+
+
+class TestAbstractMethodEnforcement:
+    """Test that abstract methods must be implemented"""
+
+    def test_incomplete_engine_cannot_be_instantiated(self):
+        """Test that incomplete engine implementations cannot be instantiated"""
+        with pytest.raises(TypeError, match="Can't instantiate abstract class"):
+            IncompleteEngine()
diff --git a/tests/test_engine/test_factory.py b/tests/test_engine/test_factory.py
new file mode 100644
index 0000000..74e5629
--- /dev/null
+++ b/tests/test_engine/test_factory.py
@@ -0,0 +1,288 @@
+from unittest.mock import MagicMock, patch
+
+import pytest
+
+from vector_search_mcp.engine import Backend, get_engine
+from vector_search_mcp.engine.base_engine import BaseEngine
+from vector_search_mcp.engine.qdrant_engine import QdrantEngine
+
+
+class TestEngineFactory:
+    """Test suite for get_engine factory function"""
+
+    def test_engine_type_enum_values(self):
+        """Test that EngineType enum has expected values"""
+        assert Backend.QDRANT == "qdrant"
+        assert len(Backend) == 2  # QDRANT and COSMOS engine types
+
+    def test_get_engine_qdrant(self):
+        """Test get_engine returns QdrantEngine for QDRANT type"""
+        with (
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.Settings"
+            ) as mock_settings_class,
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"
+            ) as mock_client_class,
+        ):
+            # Setup mocks
+            mock_settings = MagicMock()
+            mock_settings.url = "http://localhost:6333"
+            mock_settings.api_key = "test_key"
+            mock_settings_class.return_value = mock_settings
+
+            mock_client = MagicMock()
+            mock_client_class.return_value = mock_client
+
+            # Test factory function
+            engine = get_engine(Backend.QDRANT)
+
+            # Verify return type
+            assert isinstance(engine, QdrantEngine)
+            assert isinstance(engine, BaseEngine)
+
+            # Verify initialization was called correctly
+            mock_settings_class.assert_called_once()
+            mock_client_class.assert_called_once_with(
+                url=mock_settings.url, api_key=mock_settings.api_key
+            )
+
+    def test_get_engine_invalid_type(self):
+        """Test get_engine raises ValueError for unknown engine type"""
+        # Create an invalid engine type (bypassing enum validation)
+        invalid_type = "invalid_engine"
+
+        with pytest.raises(ValueError, match="Unknown engine type: invalid_engine"):
+            # We need to cast to bypass type checking
+            get_engine(invalid_type)  # type: ignore
+
+    def test_get_engine_typing_literal_qdrant(self):
+        """Test that get_engine with literal QDRANT returns correct type"""
+        with (
+            patch("vector_search_mcp.engine.qdrant_engine.Settings"),
+            patch("vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"),
+        ):
+            # When using literal Backend.QDRANT, mypy should know it's QdrantEngine
+            engine = get_engine(Backend.QDRANT)
+
+            # Runtime verification that it's the correct type
+            assert type(engine).__name__ == "QdrantEngine"
+            assert hasattr(engine, "client")  # QdrantEngine specific attribute
+            assert hasattr(engine, "settings")  # QdrantEngine specific attribute
+
+    def test_get_engine_typing_variable(self):
+        """Test that get_engine with variable returns BaseEngine type"""
+        with (
+            patch("vector_search_mcp.engine.qdrant_engine.Settings"),
+            patch("vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"),
+        ):
+            # When using a variable, mypy should see it as BaseEngine
+            engine_type: Backend = Backend.QDRANT
+            engine = get_engine(engine_type)
+
+            # Runtime verification - it's still a QdrantEngine but typed as BaseEngine
+            assert isinstance(engine, BaseEngine)
+            assert isinstance(engine, QdrantEngine)
+
+    def test_get_engine_uses_cache(self):
+        """Test that get_engine uses cache and returns same instances"""
+        with (
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.Settings"
+            ) as mock_settings_class,
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"
+            ) as mock_client_class,
+        ):
+            # Setup mocks
+            mock_settings_class.return_value = MagicMock()
+            mock_client_class.return_value = MagicMock()
+
+            # Create multiple engines
+            engine1 = get_engine(Backend.QDRANT)
+            engine2 = get_engine(Backend.QDRANT)
+
+            # Verify they are the same instance due to @cache decorator
+            assert engine1 is engine2
+            assert id(engine1) == id(engine2)
+
+            # But they are the same type
+            assert type(engine1) is type(engine2)
+            assert isinstance(engine1, QdrantEngine)
+            assert isinstance(engine2, QdrantEngine)
+
+            # Verify initialization was called only once due to caching
+            mock_settings_class.assert_called_once()
+            mock_client_class.assert_called_once()
+
+    def test_engine_type_string_values(self):
+        """Test EngineType string representations"""
+        assert str(Backend.QDRANT) == "qdrant"
+        assert str(Backend.COSMOS) == "cosmos"
+
+        # Test that it can be used in string contexts
+        engine_name = f"engine_{Backend.QDRANT}"
+        assert engine_name == "engine_qdrant"
+
+    def test_engine_type_iteration(self):
+        """Test that EngineType can be iterated over"""
+        engine_types = list(Backend)
+        assert len(engine_types) == 2
+        assert Backend.QDRANT in engine_types
+        assert Backend.COSMOS in engine_types
+
+    def test_engine_factory_integration(self):
+        """Test complete factory integration with engine functionality"""
+        with (
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.Settings"
+            ) as mock_settings_class,
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"
+            ) as mock_client_class,
+        ):
+            # Setup mocks
+            mock_settings = MagicMock()
+            mock_settings_class.return_value = mock_settings
+            mock_client_class.return_value = MagicMock()
+
+            # Create engine through factory
+            engine = get_engine(Backend.QDRANT)
+
+            # Verify engine has all required methods from BaseEngine
+            assert hasattr(engine, "transform_conditions")
+            assert hasattr(engine, "transform_response")
+            assert hasattr(engine, "run_similarity_query")
+            assert hasattr(engine, "semantic_search")
+
+            # Verify methods are callable
+            assert callable(engine.transform_conditions)
+            assert callable(engine.transform_response)
+            assert callable(engine.run_similarity_query)
+            assert callable(engine.semantic_search)
+
+    def test_future_engine_extensibility(self):
+        """Test structure supports future engine additions"""
+        # Verify that EngineType is a StrEnum and can be extended
+        assert issubclass(Backend, str)
+
+        # Verify the factory function structure can handle new engines
+        # (This is more of a design verification)
+        import inspect
+
+        sig = inspect.signature(get_engine)
+
+        # Should take Backend and return BaseEngine
+        params = list(sig.parameters.values())
+        assert len(params) == 1
+        assert params[0].name == "backend"
+
+
+class TestEngineTypeEnum:
+    """Test suite specifically for Backend enum"""
+
+    def test_engine_type_is_str_enum(self):
+        """Test that Backend is a StrEnum"""
+        from enum import StrEnum
+
+        assert issubclass(Backend, StrEnum)
+
+        # Should behave like strings
+        assert Backend.QDRANT == "qdrant"
+        assert f"{Backend.QDRANT}" == "qdrant"
+
+    def test_engine_type_comparison(self):
+        """Test EngineType comparison operations"""
+        # Should equal string value
+        assert Backend.QDRANT == "qdrant"
+
+        # Should not equal other strings
+        assert Backend.QDRANT != "other"
+        assert Backend.QDRANT != "QDRANT"  # Case sensitive
+
+    def test_engine_type_in_collections(self):
+        """Test EngineType works in collections"""
+        engine_list = [Backend.QDRANT]
+        assert Backend.QDRANT in engine_list
+        assert "qdrant" in engine_list  # StrEnum benefit
+
+        engine_set = {Backend.QDRANT}
+        assert Backend.QDRANT in engine_set
+
+    def test_engine_type_json_serializable(self):
+        """Test that Backend can be JSON serialized"""
+        import json
+
+        data = {"engine": Backend.QDRANT}
+        json_str = json.dumps(data, default=str)
+        assert '"engine": "qdrant"' in json_str
+
+    def test_engine_type_immutable(self):
+        """Test that Backend values cannot be modified"""
+        original_value = Backend.QDRANT
+
+        # Enum values should be immutable
+        with pytest.raises(AttributeError):
+            Backend.QDRANT = "modified"  # type: ignore
+
+        # Original should be unchanged
+        assert Backend.QDRANT == original_value
+
+
+class TestEngineFactoryErrorHandling:
+    """Test suite for error handling in engine factory"""
+
+    def test_none_backend_type(self):
+        """Test get_engine with None raises appropriate error"""
+        with pytest.raises((TypeError, ValueError)):
+            get_engine(None)  # type: ignore
+
+    def test_empty_string_backend_type(self):
+        """Test get_engine with empty string"""
+        with pytest.raises(ValueError, match="Unknown engine type"):
+            get_engine("")  # type: ignore
+
+    def test_numeric_backend_type(self):
+        """Test get_engine with numeric input"""
+        with pytest.raises((TypeError, ValueError)):
+            get_engine(123)  # type: ignore
+
+    def test_boolean_backend_type(self):
+        """Test get_engine with boolean input"""
+        with pytest.raises((TypeError, ValueError)):
+            get_engine(True)  # type: ignore
+
+    def test_get_engine_cosmos_not_implemented(self):
+        """Test that COSMOS engine raises NotImplementedError"""
+        with pytest.raises(
+            NotImplementedError, match="Cosmos engine is not implemented yet"
+        ):
+            get_engine(Backend.COSMOS)
+
+    def test_engine_initialization_failure(self):
+        """Test handling of engine initialization failures"""
+        with (
+            patch("vector_search_mcp.engine.qdrant_engine.Settings") as mock_settings,
+            patch("vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"),
+        ):
+            # Make Settings initialization raise an exception
+            mock_settings.side_effect = Exception("Settings initialization failed")
+
+            with pytest.raises(Exception, match="Settings initialization failed"):
+                get_engine(Backend.QDRANT)
+
+    def test_case_sensitive_backend_type(self):
+        """Test that backend type matching is case sensitive"""
+        with pytest.raises(ValueError, match="Unknown engine type"):
+            get_engine("QDRANT")  # type: ignore
+
+        with pytest.raises(ValueError, match="Unknown engine type"):
+            get_engine("Qdrant")  # type: ignore
+
+    def test_whitespace_backend_type(self):
+        """Test backend type with whitespace"""
+        with pytest.raises(ValueError, match="Unknown engine type"):
+            get_engine(" qdrant ")  # type: ignore
+
+        with pytest.raises(ValueError, match="Unknown engine type"):
+            get_engine("\tqdrant\n")  # type: ignore
diff --git a/tests/test_engine/test_integration.py b/tests/test_engine/test_integration.py
new file mode 100644
index 0000000..234647e
--- /dev/null
+++ b/tests/test_engine/test_integration.py
@@ -0,0 +1,361 @@
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+from qdrant_client import models
+
+from vector_search_mcp.engine import Backend, get_engine
+from vector_search_mcp.models import Match, MatchAny, MatchExclude, SearchRow
+
+
+class TestEngineIntegration:
+    """Integration tests for the complete engine workflow"""
+
+    @pytest.fixture
+    def mock_complete_engine_setup(self):
+        """Setup complete mocked engine environment"""
+        with (
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.Settings"
+            ) as mock_settings_class,
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"
+            ) as mock_client_class,
+        ):
+            # Setup settings
+            mock_settings = MagicMock()
+            mock_settings.url = "http://localhost:6333"
+            mock_settings.api_key = "test_api_key"
+            mock_settings_class.return_value = mock_settings
+
+            # Setup client with realistic response
+            mock_client = AsyncMock()
+            mock_client.search.return_value = [
+                models.ScoredPoint(
+                    id="doc_1",
+                    score=0.95,
+                    payload={
+                        "text": "Advanced Python programming techniques for data science",
+                        "category": "programming",
+                        "language": "python",
+                        "difficulty": "advanced",
+                        "tags": ["python", "data-science", "machine-learning"],
+                    },
+                    version=1,
+                ),
+                models.ScoredPoint(
+                    id="doc_2",
+                    score=0.87,
+                    payload={
+                        "text": "Rust systems programming for performance-critical applications",
+                        "category": "programming",
+                        "language": "rust",
+                        "difficulty": "intermediate",
+                        "tags": ["rust", "systems", "performance"],
+                    },
+                    version=1,
+                ),
+                models.ScoredPoint(
+                    id="doc_3",
+                    score=0.78,
+                    payload={
+                        "text": "Introduction to machine learning with Python",
+                        "category": "programming",
+                        "language": "python",
+                        "difficulty": "beginner",
+                        "tags": ["python", "machine-learning", "tutorial"],
+                    },
+                    version=1,
+                ),
+            ]
+            mock_client_class.return_value = mock_client
+
+            yield {
+                "settings": mock_settings,
+                "client": mock_client,
+                "settings_class": mock_settings_class,
+                "client_class": mock_client_class,
+            }
+
+    @pytest.mark.asyncio
+    async def test_complete_semantic_search_workflow(self, mock_complete_engine_setup):
+        """Test the complete workflow from factory to results"""
+        mocks = mock_complete_engine_setup
+
+        # 1. Create engine through factory
+        engine = get_engine(Backend.QDRANT)
+
+        # 2. Prepare search parameters
+        query_vector = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8]
+        collection_name = "programming_docs"
+        search_conditions = [
+            Match(key="category", value="programming"),
+            MatchAny(key="language", any=["python", "rust"]),
+            MatchExclude(key="difficulty", exclude=["expert"]),
+        ]
+
+        # 3. Execute semantic search
+        results = await engine.semantic_search(
+            embedding=query_vector,
+            collection=collection_name,
+            limit=5,
+            conditions=search_conditions,
+            threshold=0.7,
+        )
+
+        # 4. Verify the complete flow
+
+        # Check that client.search was called with correct parameters
+        client_mock = mocks["client"]
+        client_mock.search.assert_called_once()
+
+        call_args = client_mock.search.call_args
+        assert call_args[1]["collection_name"] == collection_name
+        assert call_args[1]["query_vector"] == query_vector
+        assert call_args[1]["limit"] == 5
+        assert call_args[1]["score_threshold"] == 0.7
+        assert call_args[1]["with_payload"] is True
+        assert call_args[1]["with_vectors"] is False
+
+        # Verify conditions were transformed to Qdrant filter
+        qdrant_filter = call_args[1]["query_filter"]
+        assert isinstance(qdrant_filter, models.Filter)
+        assert len(qdrant_filter.must) == 3
+
+        # Check individual conditions
+        conditions = qdrant_filter.must
+
+        # Match condition
+        match_condition = next(c for c in conditions if c.key == "category")
+        assert isinstance(match_condition.match, models.MatchValue)
+        assert match_condition.match.value == "programming"
+
+        # MatchAny condition
+        match_any_condition = next(c for c in conditions if c.key == "language")
+        assert isinstance(match_any_condition.match, models.MatchAny)
+        assert match_any_condition.match.any == ["python", "rust"]
+
+        # MatchExclude condition
+        match_exclude_condition = next(c for c in conditions if c.key == "difficulty")
+        assert isinstance(match_exclude_condition.match, models.MatchExcept)
+
+        # 5. Verify results transformation
+        assert isinstance(results, list)
+        assert len(results) == 3
+        assert all(isinstance(result, SearchRow) for result in results)
+
+        # Check first result
+        assert results[0].chunk_id == "doc_1"
+        assert results[0].score == 0.95
+        assert (
+            results[0].payload["text"]
+            == "Advanced Python programming techniques for data science"
+        )
+        assert results[0].payload["category"] == "programming"
+
+        # Check second result
+        assert results[1].chunk_id == "doc_2"
+        assert results[1].score == 0.87
+        assert results[1].payload["language"] == "rust"
+
+        # Check third result
+        assert results[2].chunk_id == "doc_3"
+        assert results[2].score == 0.78
+        assert results[2].payload["difficulty"] == "beginner"
+
+    @pytest.mark.asyncio
+    async def test_search_with_no_conditions(self, mock_complete_engine_setup):
+        """Test semantic search without any conditions"""
+        engine = get_engine(Backend.QDRANT)
+
+        results = await engine.semantic_search(
+            embedding=[0.1, 0.2, 0.3], collection="test_collection"
+        )
+
+        # Verify no filter was applied
+        client_mock = mock_complete_engine_setup["client"]
+        call_args = client_mock.search.call_args
+        assert call_args[1]["query_filter"] is None
+
+        # Results should still be transformed
+        assert len(results) == 3
+        assert all(isinstance(result, SearchRow) for result in results)
+
+    @pytest.mark.asyncio
+    async def test_search_with_empty_conditions(self, mock_complete_engine_setup):
+        """Test semantic search with empty conditions list"""
+        engine = get_engine(Backend.QDRANT)
+
+        results = await engine.semantic_search(
+            embedding=[0.1, 0.2, 0.3], collection="test_collection", conditions=[]
+        )
+
+        # Verify no filter was applied
+        client_mock = mock_complete_engine_setup["client"]
+        call_args = client_mock.search.call_args
+        assert call_args[1]["query_filter"] is None
+
+        assert len(results) == 3
+
+    @pytest.mark.asyncio
+    async def test_search_filters_null_payloads(self, mock_complete_engine_setup):
+        """Test that results with null payloads are filtered out"""
+        # Override the mock response to include null payload
+        client_mock = mock_complete_engine_setup["client"]
+        client_mock.search.return_value = [
+            models.ScoredPoint(
+                id="valid_1",
+                score=0.95,
+                payload={"text": "Valid document"},
+                version=1,
+            ),
+            models.ScoredPoint(
+                id="invalid",
+                score=0.90,
+                payload=None,  # This should be filtered out
+                version=1,
+            ),
+            models.ScoredPoint(
+                id="valid_2",
+                score=0.85,
+                payload={"text": "Another valid document"},
+                version=1,
+            ),
+        ]
+
+        engine = get_engine(Backend.QDRANT)
+        results = await engine.semantic_search(
+            embedding=[0.1, 0.2, 0.3], collection="test_collection"
+        )
+
+        # Should only have 2 results (null payload filtered out)
+        assert len(results) == 2
+        assert results[0].chunk_id == "valid_1"
+        assert results[1].chunk_id == "valid_2"
+
+    @pytest.mark.asyncio
+    async def test_error_propagation_from_client(self, mock_complete_engine_setup):
+        """Test that client errors are properly propagated"""
+        # Make the client raise an exception
+        client_mock = mock_complete_engine_setup["client"]
+        client_mock.search.side_effect = Exception("Qdrant connection timeout")
+
+        engine = get_engine(Backend.QDRANT)
+
+        with pytest.raises(Exception, match="Qdrant connection timeout"):
+            await engine.semantic_search(
+                embedding=[0.1, 0.2, 0.3], collection="test_collection"
+            )
+
+    @pytest.mark.asyncio
+    async def test_search_with_named_vector(self, mock_complete_engine_setup):
+        """Test semantic search with NamedVector instead of regular vector"""
+        engine = get_engine(Backend.QDRANT)
+
+        named_vector = models.NamedVector(
+            name="text_embedding", vector=[0.1, 0.2, 0.3, 0.4, 0.5]
+        )
+
+        results = await engine.semantic_search(
+            embedding=named_vector,  # type: ignore - Testing duck typing
+            collection="test_collection",
+        )
+
+        # Verify named vector was passed through
+        client_mock = mock_complete_engine_setup["client"]
+        call_args = client_mock.search.call_args
+        assert call_args[1]["query_vector"] == named_vector
+
+        assert len(results) == 3
+
+    @pytest.mark.asyncio
+    async def test_search_parameter_defaults(self, mock_complete_engine_setup):
+        """Test that default parameters are applied correctly"""
+        engine = get_engine(Backend.QDRANT)
+
+        await engine.semantic_search(
+            embedding=[0.1, 0.2, 0.3], collection="test_collection"
+        )
+
+        client_mock = mock_complete_engine_setup["client"]
+        call_args = client_mock.search.call_args
+
+        # Check defaults
+        assert call_args[1]["limit"] == 10  # default limit
+        assert call_args[1]["score_threshold"] is None  # default threshold
+        assert call_args[1]["query_filter"] is None  # default conditions
+        assert call_args[1]["with_payload"] is True
+        assert call_args[1]["with_vectors"] is False
+
+    @pytest.mark.asyncio
+    async def test_multiple_engine_instances_independence(
+        self, mock_complete_engine_setup
+    ):
+        """Test that multiple engine instances work independently"""
+        # Create two engines
+        engine1 = get_engine(Backend.QDRANT)
+        engine2 = get_engine(Backend.QDRANT)
+
+        # Verify they are the same instance due to caching
+        assert engine1 is engine2
+
+        # Both should work with the same instance
+        results1 = await engine1.semantic_search(
+            embedding=[0.1, 0.2, 0.3], collection="collection1"
+        )
+
+        results2 = await engine2.semantic_search(
+            embedding=[0.4, 0.5, 0.6], collection="collection2"
+        )
+
+        assert len(results1) == 3
+        assert len(results2) == 3
+
+        # Verify client was called twice (same instance, multiple calls)
+        client_mock = mock_complete_engine_setup["client"]
+        assert client_mock.search.call_count == 2
+
+    @pytest.mark.asyncio
+    async def test_large_result_set_handling(self, mock_complete_engine_setup):
+        """Test handling of large result sets"""
+        # Create a large mock response
+        large_response = []
+        for i in range(100):
+            large_response.append(
+                models.ScoredPoint(
+                    id=f"doc_{i}",
+                    score=0.9 - (i * 0.001),  # Decreasing scores
+                    payload={"text": f"Document {i}", "index": i},
+                    version=1,
+                )
+            )
+
+        client_mock = mock_complete_engine_setup["client"]
+        client_mock.search.return_value = large_response
+
+        engine = get_engine(Backend.QDRANT)
+        results = await engine.semantic_search(
+            embedding=[0.1, 0.2, 0.3], collection="large_collection", limit=100
+        )
+
+        # Should handle all 100 results
+        assert len(results) == 100
+        assert results[0].chunk_id == "doc_0"
+        assert results[0].score == 0.9
+        assert results[99].chunk_id == "doc_99"
+        assert results[99].score == 0.801  # 0.9 - (99 * 0.001)
+
+    def test_engine_type_consistency(self):
+        """Test that engine types are consistent across multiple calls"""
+        with (
+            patch("vector_search_mcp.engine.qdrant_engine.Settings"),
+            patch("vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"),
+        ):
+            engines = [get_engine(Backend.QDRANT) for _ in range(5)]
+
+            # All should be the same instance due to caching
+            assert all(engine is engines[0] for engine in engines)
+
+            # All should be QdrantEngine instances
+            from vector_search_mcp.engine.qdrant_engine import QdrantEngine
+
+            assert all(isinstance(engine, QdrantEngine) for engine in engines)
diff --git a/tests/test_engine/test_qdrant_engine.py b/tests/test_engine/test_qdrant_engine.py
new file mode 100644
index 0000000..656fb8e
--- /dev/null
+++ b/tests/test_engine/test_qdrant_engine.py
@@ -0,0 +1,380 @@
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+from qdrant_client import models
+from qdrant_client.models import ScoredPoint
+
+from vector_search_mcp.engine.base_engine import BaseEngine
+from vector_search_mcp.engine.qdrant_engine import QdrantEngine
+from vector_search_mcp.models import Match, MatchAny, MatchExclude, SearchRow
+
+
+class TestQdrantEngine:
+    """Test suite for QdrantEngine"""
+
+    @pytest.fixture
+    def mock_client(self):
+        """Create a mock Qdrant client"""
+        return AsyncMock()
+
+    @pytest.fixture
+    def mock_settings(self):
+        """Create mock settings"""
+        settings = MagicMock()
+        settings.url = "http://localhost:6333"
+        settings.api_key = "test_api_key"
+        return settings
+
+    @pytest.fixture
+    def qdrant_engine(self, mock_client, mock_settings):
+        """Create a QdrantEngine instance with mocked dependencies"""
+        with (
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.Settings"
+            ) as mock_settings_class,
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"
+            ) as mock_client_class,
+        ):
+            mock_settings_class.return_value = mock_settings
+            mock_client_class.return_value = mock_client
+
+            engine = QdrantEngine()
+            engine.client = mock_client  # Ensure we use our mock
+            return engine
+
+    def test_inheritance(self, qdrant_engine):
+        """Test that QdrantEngine properly inherits from BaseEngine"""
+        assert isinstance(qdrant_engine, BaseEngine)
+        assert isinstance(qdrant_engine, QdrantEngine)
+
+    def test_typing_parameters(self):
+        """Test that QdrantEngine has correct generic type parameters"""
+        # QdrantEngine should be BaseEngine[list[models.ScoredPoint], models.Filter]
+        # This is verified by the type checker, but we can test the methods exist
+        assert hasattr(QdrantEngine, "transform_conditions")
+        assert hasattr(QdrantEngine, "transform_response")
+        assert hasattr(QdrantEngine, "run_similarity_query")
+
+    def test_transform_conditions_none(self, qdrant_engine):
+        """Test transform_conditions with None input"""
+        result = qdrant_engine.transform_conditions(None)
+        assert result is None
+
+    def test_transform_conditions_empty_list(self, qdrant_engine):
+        """Test transform_conditions with empty list"""
+        result = qdrant_engine.transform_conditions([])
+        assert result is None
+
+    def test_transform_conditions_with_match(self, qdrant_engine):
+        """Test transform_conditions with Match condition"""
+        conditions = [Match(key="category", value="document")]
+
+        result = qdrant_engine.transform_conditions(conditions)
+
+        assert isinstance(result, models.Filter)
+        assert len(result.must) == 1
+
+        condition = result.must[0]
+        assert isinstance(condition, models.FieldCondition)
+        assert condition.key == "category"
+        assert isinstance(condition.match, models.MatchValue)
+        assert condition.match.value == "document"
+
+    def test_transform_conditions_with_match_any(self, qdrant_engine):
+        """Test transform_conditions with MatchAny condition"""
+        conditions = [MatchAny(key="tags", any=["python", "rust", "javascript"])]
+
+        result = qdrant_engine.transform_conditions(conditions)
+
+        assert isinstance(result, models.Filter)
+        assert len(result.must) == 1
+
+        condition = result.must[0]
+        assert isinstance(condition, models.FieldCondition)
+        assert condition.key == "tags"
+        assert isinstance(condition.match, models.MatchAny)
+        assert condition.match.any == ["python", "rust", "javascript"]
+
+    def test_transform_conditions_with_match_exclude(self, qdrant_engine):
+        """Test transform_conditions with MatchExclude condition"""
+        conditions = [MatchExclude(key="status", exclude=["deleted", "archived"])]
+
+        result = qdrant_engine.transform_conditions(conditions)
+
+        assert isinstance(result, models.Filter)
+        assert len(result.must) == 1
+
+        condition = result.must[0]
+        assert isinstance(condition, models.FieldCondition)
+        assert condition.key == "status"
+        assert isinstance(condition.match, models.MatchExcept)
+        # MatchExcept uses 'except' parameter which conflicts with Python keyword
+        assert hasattr(condition.match, "except_")
+
+    def test_transform_conditions_multiple(self, qdrant_engine):
+        """Test transform_conditions with multiple conditions"""
+        conditions = [
+            Match(key="type", value="article"),
+            MatchAny(key="language", any=["en", "es"]),
+            MatchExclude(key="status", exclude=["draft"]),
+        ]
+
+        result = qdrant_engine.transform_conditions(conditions)
+
+        assert isinstance(result, models.Filter)
+        assert len(result.must) == 3
+
+        # Verify all conditions are FieldCondition instances
+        assert all(isinstance(cond, models.FieldCondition) for cond in result.must)
+
+    def test_transform_response_empty(self, qdrant_engine):
+        """Test transform_response with empty response"""
+        response = []
+
+        result = qdrant_engine.transform_response(response)
+
+        assert isinstance(result, list)
+        assert len(result) == 0
+
+    def test_transform_response_with_scored_points(self, qdrant_engine):
+        """Test transform_response with valid ScoredPoint objects"""
+        response = [
+            ScoredPoint(
+                id=1,
+                score=0.95,
+                payload={"text": "First document", "category": "tech"},
+                version=1,
+            ),
+            ScoredPoint(
+                id=2,
+                score=0.87,
+                payload={"text": "Second document", "category": "science"},
+                version=1,
+            ),
+        ]
+
+        result = qdrant_engine.transform_response(response)
+
+        assert isinstance(result, list)
+        assert len(result) == 2
+
+        # Check first result
+        assert isinstance(result[0], SearchRow)
+        assert result[0].chunk_id == "1"
+        assert result[0].score == 0.95
+        assert result[0].payload == {"text": "First document", "category": "tech"}
+
+        # Check second result
+        assert isinstance(result[1], SearchRow)
+        assert result[1].chunk_id == "2"
+        assert result[1].score == 0.87
+        assert result[1].payload == {"text": "Second document", "category": "science"}
+
+    def test_transform_response_filters_none_payload(self, qdrant_engine):
+        """Test transform_response filters out points with None payload"""
+        response = [
+            ScoredPoint(
+                id=1, score=0.95, payload={"text": "Valid document"}, version=1
+            ),
+            ScoredPoint(
+                id=2,
+                score=0.87,
+                payload=None,  # This should be filtered out
+                version=1,
+            ),
+            ScoredPoint(
+                id=3, score=0.75, payload={"text": "Another valid document"}, version=1
+            ),
+        ]
+
+        result = qdrant_engine.transform_response(response)
+
+        assert isinstance(result, list)
+        assert len(result) == 2  # Only 2 valid results
+        assert result[0].chunk_id == "1"
+        assert result[1].chunk_id == "3"
+
+    @pytest.mark.asyncio
+    async def test_run_similarity_query_basic(self, qdrant_engine, mock_client):
+        """Test run_similarity_query with basic parameters"""
+        # Setup mock response
+        mock_response = [
+            ScoredPoint(id=1, score=0.9, payload={"text": "Test document"}, version=1)
+        ]
+        mock_client.search.return_value = mock_response
+
+        embedding = [0.1, 0.2, 0.3, 0.4, 0.5]
+        collection = "test_collection"
+
+        result = await qdrant_engine.run_similarity_query(
+            embedding=embedding, collection=collection
+        )
+
+        # Verify client.search was called with correct parameters
+        mock_client.search.assert_called_once_with(
+            collection_name=collection,
+            query_vector=embedding,
+            query_filter=None,
+            limit=10,  # default
+            with_payload=True,
+            with_vectors=False,
+            score_threshold=None,
+        )
+
+        assert result == mock_response
+
+    @pytest.mark.asyncio
+    async def test_run_similarity_query_with_all_parameters(
+        self, qdrant_engine, mock_client
+    ):
+        """Test run_similarity_query with all parameters"""
+        mock_response = []
+        mock_client.search.return_value = mock_response
+
+        embedding = [0.1, 0.2, 0.3]
+        collection = "test_collection"
+        limit = 5
+        conditions = models.Filter(must=[])
+        threshold = 0.75
+
+        result = await qdrant_engine.run_similarity_query(
+            embedding=embedding,
+            collection=collection,
+            limit=limit,
+            conditions=conditions,
+            threshold=threshold,
+        )
+
+        mock_client.search.assert_called_once_with(
+            collection_name=collection,
+            query_vector=embedding,
+            query_filter=conditions,
+            limit=limit,
+            with_payload=True,
+            with_vectors=False,
+            score_threshold=threshold,
+        )
+
+        assert result == mock_response
+
+    @pytest.mark.asyncio
+    async def test_run_similarity_query_with_named_vector(
+        self, qdrant_engine, mock_client
+    ):
+        """Test run_similarity_query with NamedVector"""
+        mock_response = []
+        mock_client.search.return_value = mock_response
+
+        named_vector = models.NamedVector(name="text", vector=[0.1, 0.2, 0.3])
+        collection = "test_collection"
+
+        result = await qdrant_engine.run_similarity_query(
+            embedding=named_vector, collection=collection
+        )
+
+        mock_client.search.assert_called_once_with(
+            collection_name=collection,
+            query_vector=named_vector,
+            query_filter=None,
+            limit=10,
+            with_payload=True,
+            with_vectors=False,
+            score_threshold=None,
+        )
+
+        assert result == mock_response
+
+    @pytest.mark.asyncio
+    async def test_semantic_search_integration(self, qdrant_engine, mock_client):
+        """Test the full semantic_search flow through QdrantEngine"""
+        # Setup mock response
+        mock_search_response = [
+            ScoredPoint(
+                id=1,
+                score=0.95,
+                payload={"text": "Python programming guide", "category": "tech"},
+                version=1,
+            ),
+            ScoredPoint(
+                id=2,
+                score=0.87,
+                payload={"text": "Rust systems programming", "category": "tech"},
+                version=1,
+            ),
+        ]
+        mock_client.search.return_value = mock_search_response
+
+        # Test data
+        vector = [0.1, 0.2, 0.3, 0.4, 0.5]
+        collection = "documents"
+        conditions = [
+            Match(key="category", value="tech"),
+            MatchAny(key="language", any=["python", "rust"]),
+        ]
+
+        result = await qdrant_engine.semantic_search(
+            embedding=vector,
+            collection=collection,
+            limit=5,
+            conditions=conditions,
+            threshold=0.8,
+        )
+
+        # Verify the search was called with transformed conditions
+        assert mock_client.search.called
+        call_args = mock_client.search.call_args
+        assert call_args[1]["collection_name"] == collection
+        assert call_args[1]["query_vector"] == vector
+        assert call_args[1]["limit"] == 5
+        assert call_args[1]["score_threshold"] == 0.8
+        assert isinstance(call_args[1]["query_filter"], models.Filter)
+
+        # Verify the response was transformed correctly
+        assert isinstance(result, list)
+        assert len(result) == 2
+        assert all(isinstance(row, SearchRow) for row in result)
+        assert result[0].chunk_id == "1"
+        assert result[0].score == 0.95
+        assert result[1].chunk_id == "2"
+        assert result[1].score == 0.87
+
+    def test_initialization_with_settings(self, mock_settings):
+        """Test QdrantEngine initialization uses settings correctly"""
+        with (
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.Settings"
+            ) as mock_settings_class,
+            patch(
+                "vector_search_mcp.engine.qdrant_engine.AsyncQdrantClient"
+            ) as mock_client_class,
+        ):
+            mock_settings_class.return_value = mock_settings
+            mock_client = AsyncMock()
+            mock_client_class.return_value = mock_client
+
+            engine = QdrantEngine()
+
+            # Verify Settings was instantiated
+            mock_settings_class.assert_called_once()
+
+            # Verify AsyncQdrantClient was created with correct parameters
+            mock_client_class.assert_called_once_with(
+                url=mock_settings.url, api_key=mock_settings.api_key
+            )
+
+            assert engine.client == mock_client
+            assert engine.settings == mock_settings
+
+    @pytest.mark.asyncio
+    async def test_client_search_exception_propagation(
+        self, qdrant_engine, mock_client
+    ):
+        """Test that exceptions from client.search are properly propagated"""
+        # Setup mock to raise an exception
+        mock_client.search.side_effect = Exception("Qdrant connection failed")
+
+        with pytest.raises(Exception, match="Qdrant connection failed"):
+            await qdrant_engine.run_similarity_query(
+                embedding=[0.1, 0.2, 0.3], collection="test_collection"
+            )