Pocket Studio Academy
HomePart 66.8

Where to go next

Free lesson6 min read·3 questions

The last lesson. What you can actually do now, five small apps worth building next, how to keep learning once a course stops teaching you, and an honest note about how much of this you already know.

Look at what is on your home screen

Three icons that were not there when you started. A dice game with an animation you wrote. A notes app with a real database underneath it. A focus timer with settings, notifications and a chart you drew by hand, pixel by pixel, on a Canvas.

You built all of that on a phone. No laptop was involved at any point.

That is not a small thing, and it is worth sitting with for a second before you read the rest of this page.

Think of it like this

Think about the day someone passes their driving test.

They can drive. Genuinely — they can pull away, read a roundabout, park between two cars, handle a hill start in the rain. Nobody handed them that; they earned it over months.

And they have also never driven on a motorway alone, never driven a car that is not the one they learned in, never had a tyre go flat on a dark road. The test was not the finish. It was the point at which somebody stopped sitting in the passenger seat.

You are exactly there. You know the controls. What comes next is roads.

What you can actually do now

Not a pep talk — a list. Everything here is something you did, not something you read about.

  • Write Kotlin. and , types, , collections, , , , and .
  • Explain what happens when someone taps an icon. The , the , the , the , and what is actually inside an .
  • Build a screen in — layout, , , theming, lists, text input and animation.
  • Structure an app that survives. , a , a , , , and a .
  • Read an error. This is the one that quietly matters most. You have met around two hundred real error messages in this course and you know that the fix is usually three lines above the scary bit.
  • Ship. Icons, empty states, , , , , and what Google Play really asks for.

Build your own thing next, and make it small

Here is the advice that decides whether people keep going: cut your idea down until it fits on one screen.

Not because your idea is too ambitious. Because a finished one-screen app teaches you more than an abandoned five-screen one, and because the second version is easy once the first exists.

Five ideas, each one buildable with exactly what you already know:

1. A habit tracker. One table, one LazyColumn, a tick box per row. This is Pocket Notes chapters 1 to 3 with different words. The whole data model is six lines:

kotlin
1@Entity
2data class Habit(
3  @PrimaryKey(autoGenerate = true) val id: Int = 0,
4  val name: String,
5  val doneToday: Boolean = false
6)

2. A spending log. An amount, a category, a date. A list, and a total at the top from sumOf. That is Lesson 1.12 and Lesson 4.5 in one small app, and you will actually use it.

3. A flashcard deck. One card at a time, tap to flip. Room holds the cards; Dice Duel chapter 4's animation does the flip.

4. A score pad for a game your family actually plays. Start from your own Dice Duel checkpoint and change the rules to theirs.

5. A "what shall we eat" picker. A list you type in, one button, random(). Buildable in an evening with Part 3 alone, and genuinely useful about twice a week.

The three apps are yours

Every checkpoint ZIP on the Checkpoint downloads page is a complete, working project. They are not read-only examples. Open one, break it, fix it, keep the version you like.

Some first changes that will each teach you something specific:

  • Dice Duel — make it best of five instead of first to a target. Add a third player. Keep a history of every roll and show it under the board.
  • Pocket Notes — add a tag field to the note. That means a Room , which is the single most useful thing in Lesson 4.5 and the one you did not have to do yet.
  • Focus Flow — make every fourth break a long one. Add a weekly total to the stats screen. Give focus sessions and breaks different colours in the ring.

Where to look things up

Courses end. Documentation does not.

is the official source, and it is better than its reputation. Three parts of it are worth bookmarking today: the Compose guides, the (free, step-by-step, genuinely good), and the when you need to know exactly what a function does.

kotlinlang.org is the language's own documentation, and it is unusually readable. The there runs code straight in a browser, which is handy on a phone when you want to test one idea without making a whole project.

Jetpack tell you what changed in each library version. Ten minutes there every couple of months keeps you current with almost no effort.

is a complete open-source app written by Google as a worked example. Reading real code is the step after finishing a course, and this is a good first one to read.

And the free tool you already have: long-press any Compose function in the Pocket Studio editor and use . You end up in the actual source of the library. It is startling the first time, and it is how you stop treating libraries as magic.

How to spot a tutorial that will waste your afternoon

Android is old enough that a lot of what you will find teaches the previous way of doing things. It is not wrong, it is just not what you learned. Four quick tells:

  • It builds screens from XML layout files and findViewById — that is the View toolkit, which predates Compose.
  • Its build file is build.gradle with single quotes, not build.gradle.kts.
  • It uses kapt where you would use .
  • It imports androidx.compose.material rather than androidx.compose.material3.

Anything written before mid-2021 predates stable Compose entirely. Check the date first.

One honest note about pace

Lesson 0.7 said the enemy was not difficulty, it was burning out. That was true then and it is truer now, because nobody is setting you lessons any more.

Twenty minutes on your own app beats a four-hour session once a month. Every time. The habit is the whole skill.

Try it in Pocket Studio

Do this now, before you close this page. Momentum after finishing something is the cheapest momentum there is, and it does not survive a night's sleep.

  1. Open Pocket Studio and tap Projects.
  2. Tap New Project and choose the empty Jetpack Compose template.
  3. Name it after one of the five ideas above, or your own. Set the package name to something that is yours — your name reversed works fine.
  4. Tap Editor and open MainActivity.kt.
  5. Find the Text(...) the template put there and replace what is inside the quotes with one sentence describing what this app will do when it is finished — something like Text("Tells me which habit I skipped today.").
  6. Tap Build, then press Run. Install it. It does nothing yet, and that is fine.
  7. Put the icon on your home screen next to Dice Duel, Pocket Notes and Focus Flow.
  8. Tomorrow, open it and add one thing. Just one.
Error Doctor4 common errors
e: file:///…/app/build.gradle.kts:32:20: Unexpected tokens (use ';' to separate expressions on the same line)
MeansYou pasted a dependency line from a tutorial that used the older Groovy build file — implementation 'group:name:1.0' — into a Kotlin one. Kotlin needs parentheses and double quotes.
FixRewrite it as implementation("group:name:1.0"). Better still, add the library to gradle/libs.versions.toml and refer to it as libs.something, the way every build file in this course does.
e: file:///…/MainActivity.kt:12:29: Unresolved reference: material3
MeansYou copied a composable into a new project and its library is not there. A fresh project only has what its template asked for, and the import line goes red because nothing on the classpath matches it.
FixOpen app/build.gradle.kts and add the missing implementation(...) line, then let Gradle sync. Compare it against a checkpoint project's dependency list — those are known-good and pinned to versions that work together.
e: file:///…/MainActivity.kt:18:22: Unresolved reference: layout
MeansR.layout.something means the tutorial you are following builds its screen from an XML layout file. A Compose project has no res/layout folder at all, so the R.layout part of the generated R class does not exist.
FixNothing is broken — you are reading a pre-Compose tutorial. Either find a Compose version of it, or translate it yourself: whatever the XML declares as a TextView or Button is a Text or Button composable in a Column.
Execution failed for task ':app:checkDebugAarMetadata'. > Dependency 'androidx.core:core-ktx:1.16.0' requires libraries and applications that depend on it to compile against version 35 or later of the Android APIs.
MeansYou added or upgraded a library that was built against a newer Android than your project compiles against. This is the most common error you will meet the first time you update something on your own.
FixRaise compileSdk in app/build.gradle.kts to the number in the message. compileSdk only controls what you are allowed to type against — Lesson 2.7 — so raising it does not change which phones can install your app. Raising targetSdk is the one that needs testing.
Recap
  • You can write , build a screen, structure an app around a and a database, read a real error, and get an installed and running — all of it from the device in your hand.
  • Build something of your own next, and cut it down to one screen. A finished small thing beats an abandoned big one, every single time.
  • The 21 checkpoint projects on the Checkpoint downloads page are yours to open, break and change. Adding a tag field to Pocket Notes is a better afternoon than any tutorial.
  • Keep learning from developer.android.com and kotlinlang.org — the guides, the and the — and read real code like .
  • Check the date on any tutorial. XML layouts and findViewById mean you are reading about the toolkit that came before the one you know.
  • Twenty minutes at a time. That is the whole trick.
  • Next: nothing — this was the last lesson, and you finished it. So: open Pocket Studio, tap New Project, and start the app you thought of somewhere around Part 3. The Glossary, the Review deck and every checkpoint project stay here for whenever you want them. Thank you for reading all of it. Now go and build something nobody asked you for.