diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
new file mode 100644
index 0000000..cf2096e
--- /dev/null
+++ b/.github/workflows/test.yml
@@ -0,0 +1,23 @@
+name: test
+
+on:
+  push:
+    branches:
+      - master
+      - main
+  pull_request:
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - uses: erlef/setup-beam@v1
+        with:
+          otp-version: "26.0.2"
+          gleam-version: "0.32.4"
+          rebar3-version: "3"
+          # elixir-version: "1.15.4"
+      - run: gleam deps download
+      - run: gleam test
+      - run: gleam format --check src test
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..170cca9
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,4 @@
+*.beam
+*.ez
+build
+erl_crash.dump
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..5c8a542
--- /dev/null
+++ b/README.md
@@ -0,0 +1,22 @@
+# glubs
+
+[![Package Version](https://img.shields.io/hexpm/v/glubs)](https://hex.pm/packages/glubs)
+[![Hex Docs](https://img.shields.io/badge/hex-docs-ffaff3)](https://hexdocs.pm/glubs/)
+
+## Quick start
+
+```sh
+gleam run   # Run the project
+gleam test  # Run the tests
+gleam shell # Run an Erlang shell
+```
+
+## Installation
+
+If available on Hex this package can be added to your Gleam project:
+
+```sh
+gleam add glubs
+```
+
+and its documentation can be found at <https://hexdocs.pm/glubs>.
diff --git a/gleam.toml b/gleam.toml
new file mode 100644
index 0000000..4dede06
--- /dev/null
+++ b/gleam.toml
@@ -0,0 +1,16 @@
+name = "glubs"
+version = "0.1.0"
+
+# Fill out these fields if you intend to generate HTML documentation or publish
+# your project to the Hex package manager.
+#
+# description = ""
+# licences = ["Apache-2.0"]
+# repository = { type = "github", user = "username", repo = "project" }
+# links = [{ title = "Website", href = "https://gleam.run" }]
+
+[dependencies]
+gleam_stdlib = "~> 0.32"
+
+[dev-dependencies]
+gleeunit = "~> 1.0"
diff --git a/manifest.toml b/manifest.toml
new file mode 100644
index 0000000..92768e9
--- /dev/null
+++ b/manifest.toml
@@ -0,0 +1,11 @@
+# This file was generated by Gleam
+# You typically do not need to edit this file
+
+packages = [
+  { name = "gleam_stdlib", version = "0.32.1", build_tools = ["gleam"], requirements = [], otp_app = "gleam_stdlib", source = "hex", outer_checksum = "ABF00CDCCB66FABBCE351A50060964C4ACE798F95A0D78622C8A7DC838792577" },
+  { name = "gleeunit", version = "1.0.0", build_tools = ["gleam"], requirements = ["gleam_stdlib"], otp_app = "gleeunit", source = "hex", outer_checksum = "D3682ED8C5F9CAE1C928F2506DE91625588CC752495988CBE0F5653A42A6F334" },
+]
+
+[requirements]
+gleam_stdlib = { version = "~> 0.32" }
+gleeunit = { version = "~> 1.0" }
diff --git a/src/glubs.gleam b/src/glubs.gleam
new file mode 100644
index 0000000..8b13789
--- /dev/null
+++ b/src/glubs.gleam
@@ -0,0 +1 @@
+
diff --git a/src/glubs/webvtt.gleam b/src/glubs/webvtt.gleam
new file mode 100644
index 0000000..2c61ec6
--- /dev/null
+++ b/src/glubs/webvtt.gleam
@@ -0,0 +1,120 @@
+import gleam/option.{None, Some}
+import gleam/string
+import gleam/result
+import gleam/list
+import gleam/int
+
+pub type Token {
+  StartTag(
+    tag: String,
+    classes: List(String),
+    annotation: option.Option(String),
+  )
+  Text(content: String)
+  Timestamp(ms: Int)
+  EndTag(tag: String)
+}
+
+pub type TokenizationError {
+  InvalidStartToken
+  InvalidEndToken
+}
+
+pub fn tokenize(payload: String) -> Result(List(Token), TokenizationError) {
+  payload
+  |> do_tokenize([])
+  |> result.map(list.reverse)
+}
+
+fn do_tokenize(
+  payload: String,
+  acc: List(Token),
+) -> Result(List(Token), TokenizationError) {
+  case payload {
+    "" -> Ok(acc)
+    "</" <> rest -> {
+      case string.split_once(rest, on: ">") {
+        Ok(#(tag, rest)) -> {
+          do_tokenize(rest, [EndTag(tag: tag), ..acc])
+        }
+        Error(Nil) -> {
+          Error(InvalidEndToken)
+        }
+      }
+    }
+    "<" <> rest -> {
+      case string.split_once(rest, on: ">") {
+        Ok(#(tag, rest)) -> {
+          case parse_timestamp(tag) {
+            Ok(ts) -> do_tokenize(rest, [Timestamp(ts), ..acc])
+            Error(_) -> do_tokenize(rest, [parse_start_tag(tag), ..acc])
+          }
+        }
+        Error(Nil) -> {
+          Error(InvalidStartToken)
+        }
+      }
+    }
+    text -> {
+      case string.split_once(text, on: "<") {
+        Ok(#(content, rest)) -> {
+          do_tokenize("<" <> rest, [Text(content), ..acc])
+        }
+        Error(Nil) -> Ok([Text(text), ..acc])
+      }
+    }
+  }
+}
+
+fn parse_start_tag(input: String) -> Token {
+  case string.split_once(input, on: " ") {
+    Ok(#(tag_and_classes, annotation)) -> {
+      let #(tag, classes) = parse_tag_and_classes(tag_and_classes)
+      StartTag(tag: tag, classes: classes, annotation: Some(annotation))
+    }
+    Error(_) -> {
+      let #(tag, classes) = parse_tag_and_classes(input)
+      StartTag(tag: tag, classes: classes, annotation: None)
+    }
+  }
+}
+
+fn parse_tag_and_classes(input: String) -> #(String, List(String)) {
+  let [tag, ..classes] = string.split(input, on: ".")
+  #(tag, classes)
+}
+
+fn parse_timestamp(input: String) -> Result(Int, Nil) {
+  case string.split(input, on: ":") {
+    [hours, minutes, seconds_and_ms] -> {
+      use hours <- result.try(int.parse(hours))
+      use minutes <- result.try(int.parse(minutes))
+      use #(seconds, ms) <- result.try(split_seconds(seconds_and_ms))
+
+      Ok({ seconds + minutes * 60 + hours * 60 * 60 } * 1000 + ms)
+    }
+
+    [minutes, seconds_and_ms] -> {
+      use minutes <- result.try(int.parse(minutes))
+      use #(seconds, ms) <- result.try(split_seconds(seconds_and_ms))
+
+      Ok({ seconds + minutes * 60 } * 1000 + ms)
+    }
+
+    [_] -> Error(Nil)
+  }
+}
+
+fn split_seconds(input: String) -> Result(#(Int, Int), Nil) {
+  case string.split_once(input, on: ".") {
+    Ok(#(seconds, ms)) -> {
+      use seconds <- result.try(int.parse(seconds))
+      use ms <- result.try(int.parse(ms))
+      Ok(#(seconds, ms))
+    }
+    Error(_) -> {
+      use seconds <- result.try(int.parse(input))
+      Ok(#(seconds, 0))
+    }
+  }
+}
diff --git a/test/glubs/webvtt_test.gleam b/test/glubs/webvtt_test.gleam
new file mode 100644
index 0000000..df404f6
--- /dev/null
+++ b/test/glubs/webvtt_test.gleam
@@ -0,0 +1,46 @@
+import gleeunit/should
+import gleam/option.{None, Some}
+import glubs/webvtt.{EndTag, StartTag, Text, Timestamp}
+
+pub fn tokenize_text_test() {
+  "Hello"
+  |> webvtt.tokenize()
+  |> should.equal(Ok([Text("Hello")]))
+}
+
+pub fn tokenize_voice_test() {
+  "<v.loud.shout Rob>Hello</v>"
+  |> webvtt.tokenize()
+  |> should.equal(Ok([
+    StartTag("v", classes: ["loud", "shout"], annotation: Some("Rob")),
+    Text("Hello"),
+    EndTag("v"),
+  ]))
+}
+
+pub fn timestamp_tag_test() {
+  "Hello <00:19.500>Phil. <01:00:00.500>How are you?"
+  |> webvtt.tokenize()
+  |> should.equal(Ok([
+    Text("Hello "),
+    Timestamp(19_500),
+    Text("Phil. "),
+    Timestamp(3_600_500),
+    Text("How are you?"),
+  ]))
+}
+
+pub fn complex_test() {
+  "<v Phil>Hi!\n<v.loud.shout Rob>Hello <i>mate!</i></v>"
+  |> webvtt.tokenize()
+  |> should.equal(Ok([
+    StartTag("v", classes: [], annotation: Some("Phil")),
+    Text("Hi!\n"),
+    StartTag("v", classes: ["loud", "shout"], annotation: Some("Rob")),
+    Text("Hello "),
+    StartTag("i", classes: [], annotation: None),
+    Text("mate!"),
+    EndTag("i"),
+    EndTag("v"),
+  ]))
+}
diff --git a/test/glubs_test.gleam b/test/glubs_test.gleam
new file mode 100644
index 0000000..ecd12ad
--- /dev/null
+++ b/test/glubs_test.gleam
@@ -0,0 +1,5 @@
+import gleeunit
+
+pub fn main() {
+  gleeunit.main()
+}