diff --git a/.travis.yml b/.travis.yml
index 587b83da..ca515e3c 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -14,8 +14,38 @@ matrix:
   include:
     - python: "3.6"
       env: TRAVIS_PYLINT=true
+    - os : osx
+      language: generic
+      python: 3.7
+      env: PYTHON=3.7
+    - os: windows
+      language: sh
+      python: 3.7
+      before_install:
+        - choco install python3
+        - export PATH="/c/Python37:/c/Python37/Scripts:$PATH"
+        - python -m pip install --upgrade pip wheel
+      script: 
+        - python -m virtualenv .env
+        - .env/Scripts/pip.exe install -e .[testing,doc]
+        - .env/Scripts/py.test.exe --cov=slash --cov-report=html tests
+  allow_failures:
+    - os: windows
 
 # command to run tests, e.g. python setup.py test
+before_install: |
+  if [ "$TRAVIS_OS_NAME" == "osx" ]; then
+    brew update
+    brew install openssl readline
+    brew outdated pyenv || brew upgrade pyenv
+    brew install pyenv-virtualenv
+    pyenv install $PYTHON
+    export PYENV_VERSION=$PYTHON
+    export PATH="/Users/travis/.pyenv/shims:${PATH}"
+    pyenv-virtualenv venv
+    source venv/bin/activate
+    python --version
+  fi
 install: 
   - pip install virtualenv
   - pip install codecov
diff --git a/azure-pipelines.yml b/azure-pipelines.yml
new file mode 100644
index 00000000..0dc73aa7
--- /dev/null
+++ b/azure-pipelines.yml
@@ -0,0 +1,49 @@
+# Python package
+# Create and test a Python package on multiple Python versions.
+# Add steps that analyze code, save the dist with the build record, publish to a PyPI-compatible index, and more:
+# https://docs.microsoft.com/azure/devops/pipelines/languages/python
+
+trigger:
+- master
+
+jobs:
+
+- job: 'Test'
+  pool:
+    vmImage: 'vs2017-win2016'
+  strategy:
+    matrix:
+      Python35:
+        python.version: '3.5'
+      Python36:
+        python.version: '3.6'
+      Python37:
+        python.version: '3.7'
+    maxParallel: 4
+
+  steps:
+  - task: UsePythonVersion@0
+    inputs:
+      versionSpec: '$(python.version)'
+      architecture: 'x64'
+
+  - script: |
+      python -m pip install --upgrade pip && pip install .[testing,doc]
+    displayName: 'Install dependencies'
+
+  - script: |
+      mkdir d:\temp && py.test tests --cov=slash --cov-report=html --cov-report=xml --junitxml=junit/test-results.xml
+    displayName: 'pytest'
+    env: {TMPDIR: 'd:\temp\'}
+ 
+  - task: PublishTestResults@2
+    condition: succeededOrFailed()
+    inputs:
+      testResultsFiles: '**/test-*.xml'
+      testRunTitle: 'Publish test results for Python $(python.version)'
+  
+  - task: PublishCodeCoverageResults@1
+    inputs:
+      codeCoverageTool: Cobertura
+      summaryFileLocation: '$(System.DefaultWorkingDirectory)/**/coverage.xml'
+      reportDirectory: '$(System.DefaultWorkingDirectory)/**/htmlcov'
diff --git a/tests/test_logging.py b/tests/test_logging.py
index 26e2c819..65f41dc7 100644
--- a/tests/test_logging.py
+++ b/tests/test_logging.py
@@ -1,6 +1,7 @@
 # pylint: disable=unused-argument,redefined-outer-name
 import functools
 import os
+import sys
 
 import logbook
 import brotli
@@ -45,7 +46,7 @@ def test_with_log_error():
     assert [res.message for res in result.get_errors()] == expected_error_messages
     assert [rec.message for rec in test_handler.records] == [message, non_capture_message]
 
-
+@pytest.mark.skipif(sys.platform == 'win32', reason="does not run on windows")
 def test_last_session_symlinks(files_dir, links_dir, session):
 
     test_log_file = files_dir.join(
@@ -154,7 +155,7 @@ def test_log_symlinks_without_root_path(suite, config_override, symlink_name):
     config_override('log.{}'.format(symlink_name), 'some/subdir')
     assert suite.run().ok()
 
-
+@pytest.mark.skipif(sys.platform == 'win32', reason="does not run on windows")
 def test_last_test_not_overriden_by_stop_on_error(links_dir, suite):
     failed_test = suite[4]
     failed_test.when_run.fail()
@@ -192,7 +193,7 @@ def test_result_log_links(files_dir, session):
         assert result.get_log_path() is not None
         assert result.get_log_path().startswith(str(files_dir))
 
-
+@pytest.mark.skipif(sys.platform == 'win32', reason="does not run on windows")
 def test_last_failed(suite, links_dir):
     suite[-5].when_run.fail()
     last_failed = suite[-2]
@@ -216,6 +217,7 @@ def test_errors_log_for_test(suite, suite_test, errors_log_path, logs_dir):
         lines = f.read().splitlines()
         assert error_line in lines
 
+@pytest.mark.skipif(sys.platform == 'win32', reason="does not run on windows")
 @pytest.mark.parametrize('should_keep_failed_tests', [True, False])
 def test_logs_deletion(suite, suite_test, errors_log_path, logs_dir, config_override, should_keep_failed_tests):
     config_override('log.cleanup.enabled', True)
diff --git a/tests/test_parallel.py b/tests/test_parallel.py
index 8ce3b286..bf418979 100644
--- a/tests/test_parallel.py
+++ b/tests/test_parallel.py
@@ -1,6 +1,8 @@
 import slash
 import os
 import signal
+import sys
+
 from .utils.suite_writer import Suite
 from slash.resuming import get_tests_from_previous_session
 from slash.exceptions import InteractiveParallelNotAllowed, ParallelTimeout
@@ -12,6 +14,9 @@
 import tempfile
 import pytest
 
+if sys.platform.startswith("win"):
+    pytest.skip("does not run on windows", allow_module_level=True)
+
 @pytest.fixture(scope='module', autouse=True)
 def no_parallel_user_config(request):
     tmpdir = tempfile.mkdtemp()
diff --git a/tests/test_suite_files.py b/tests/test_suite_files.py
index bf39af89..98787354 100644
--- a/tests/test_suite_files.py
+++ b/tests/test_suite_files.py
@@ -24,8 +24,7 @@ def test_iter_suite_paths_files_relpath(filename, paths):
             f.write(relpath)
             f.write('\n')
 
-    assert [p for p, _ in suite_files.iter_suite_file_paths([filename])] == [os.path.abspath(p) for p in paths]
-
+    assert [os.path.normcase(p) for p, _ in suite_files.iter_suite_file_paths([filename])] == [os.path.normcase(os.path.abspath(p)) for p in paths]
 
 def test_suite_files(suite, suite_test, suite_file):  # pylint: disable=unused-argument
     suite.run(args=[])
diff --git a/tests/test_test_loading.py b/tests/test_test_loading.py
index 1c6d6b5a..34bd638c 100644
--- a/tests/test_test_loading.py
+++ b/tests/test_test_loading.py
@@ -54,7 +54,7 @@ def test_dir(tmpdir, names, indices):
 
 @pytest.fixture
 def names():
-    return ['a/test_a_b.py', 'a/test_b_a.py', 'b/test_a_a.py', 'b/test_a_b.py']
+    return [os.path.normpath(p) for p in ['a/test_a_b.py', 'a/test_b_a.py', 'b/test_a_a.py', 'b/test_a_b.py']]
 
 def _sorted_by_indices(items, indices):
     returned = [None for _ in items]
diff --git a/tests/test_warnings.py b/tests/test_warnings.py
index dd3ff4ec..f0f80cdc 100644
--- a/tests/test_warnings.py
+++ b/tests/test_warnings.py
@@ -1,4 +1,5 @@
 # pylint: disable=unused-variable,redefined-outer-name
+import os
 import collections
 import warnings
 from contextlib import contextmanager
@@ -44,7 +45,7 @@ def warning_added(warning):
     assert isinstance(w.filename, str)
     assert w.lineno
     assert w.filename
-    assert w.filename.rsplit('/', 1)[-1] == suite_test.file.get_relative_path()
+    assert w.filename.rsplit(os.path.sep, 1)[-1] == suite_test.file.get_relative_path()
     warning_type = w.details['type']
     assert isinstance(warning_type, str)
     assert warning_type == 'LogbookWarning'
diff --git a/tests/utils/__init__.py b/tests/utils/__init__.py
index fb78b9dc..8b3ac42a 100644
--- a/tests/utils/__init__.py
+++ b/tests/utils/__init__.py
@@ -43,7 +43,7 @@ def override_config(self, path, value):
 
     def get_new_path(self):
         returned = tempfile.mkdtemp()
-        self.addCleanup(shutil.rmtree, returned)
+        self.addCleanup(shutil.rmtree, returned, ignore_errors=True)
         return returned
 
     _forge = None