diff --git a/pyproject.toml b/pyproject.toml
index 891f702..ee5c967 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -14,8 +14,8 @@
 # to PyPI. See https://packaging.python.org/en/latest/tutorials/packaging-projects/
 
 # To build RenderCV, we need to specify which build package we want to use. There are
-# many build packages like `setuptools`, `flit`, `poetry`, `hatch`, etc. We will use
-# `hatch`.
+# many build packages like `setuptools`, `flit`, `poetry`, `hatchling`, etc. We will use
+# `hatchling`.
 requires = ["hatchling"] # Our dependency to build RenderCV
 
 # Python has a standard object format called build-backend object. Python standard asks
@@ -37,11 +37,9 @@ path = "rendercv/__init__.py"
 include = ["/README.md", "/rendercv"]
 
 [project]
-# Under the `project` section, we specify the metadata about RenderCV.e will use
-# hatchling to generate the version number of RenderCV. It will go to the `path` below
-# and get the version number from there.
+# Under the `project` section, we specify the metadata about RenderCV.
 name = 'rendercv'
-description = 'LaTeX CV generator from a YAML/JSON input file'
+description = 'LaTeX CV generator engine from a YAML input file'
 dynamic = [
     "version",
 ] # We will use hatchling to generate the version number                                         
@@ -51,16 +49,14 @@ readme = "README.md"
 # RenderCV depends on these packages. They will be installed automatically when RenderCV
 # is installed:
 dependencies = [
-    'annotated-types==0.6.0',
-    'Jinja2==3.1.3',
-    'phonenumbers==8.13.28',
-    'pydantic==2.5.3',
-    'pydantic-extra-types==2.4.1',
-    'typing_extensions==4.9.0',
-    'ruamel.yaml==0.18.5',
-    'email-validator==2.1.0.post1',
-    'typer[all]==0.9.0',
-    "markdown==3.5.2",
+    'Jinja2==3.1.3',                # to generate LaTeX and Markdown files
+    'phonenumbers==8.13.30',        # to validate phone numbers
+    'email-validator==2.1.0.post1', # to validate email addresses
+    'pydantic==2.6.1',              # to validate and parse the input file
+    'pydantic-extra-types==2.5.0',  # to validate some extra types
+    'ruamel.yaml==0.18.6',          # to parse YAML files
+    'typer[all]==0.9.0',            # to create the command-line interface
+    "markdown==3.5.2",              # to convert Markdown to HTML
 ]
 classifiers = [
     "Intended Audience :: Science/Research",
@@ -96,21 +92,20 @@ rendercv = 'rendercv.__main__:app'
 # documentation of RenderCV, we need to install some packages. However, not all the
 # users of RenderCV will build the documentation, so these are optional dependencies.
 
-# The documentation of RenderCV is built using MkDocs.
-building_docs = ["mkdocs-material==9.5.4", "mkdocstrings-python==1.8.0"]
-
-# For tests, RenderCV uses pytest. Coverage package is used to generate coverage
-# reports.
-testing = [
-    "coverage==7.4.0",
-    "pytest==7.4.4",
-    "time-machine==2.13.0",
-    "pypdf==4.0.1",
+docs = [
+    "mkdocs-material==9.5.9",     # to build docs
+    "mkdocstrings-python==1.8.0", # to build reference documentation from docstrings
+]
+tests = [
+    "pytest==8.0.1",        # to run the tests
+    "coverage==7.4.1",      # to generate coverage reports
+    "time-machine==2.13.0", # to select an arbitrary date and time for testing
+    "pypdf==4.0.2",         # to read PDF files
+]
+dev = [
+    "ruff==0.2.2",   # to lint the code
+    "black==24.2.0", # to format the code
 ]
-
-# Ruff is a linter (a tool to flag stylistic errors, bugs, suspicious constructs, etc.)
-# Black is a code formatter.
-development = ["ruff==0.1.13", "black==24.1.1"]
 
 
 # RenderCV uses different tools to check the code quality, format the code, build the
@@ -120,10 +115,12 @@ development = ["ruff==0.1.13", "black==24.1.1"]
 # file automatically.
 
 [tool.black]
-line-length = 88                                # The maximum line length
-preview = true
-enable-unstable-feature = ["string_processing"]
+line-length = 88 # maximum line length
+preview = true # to allow enable-unstable-feature
+enable-unstable-feature = [
+    "string_processing",
+] # breaking strings into multiple lines
 
 [tool.coverage.run]
 source = ['rendercv']
-relative_files = true
+relative_files = true # use relative paths instead of absolute paths, this is useful for combining coverage reports from different OSes