Speaker

Akniyet Arysbayev

Akniyet Arysbayev

Delivery Hero (Berlin), Senior software engineer at App Performance team

Berlin, Germany

Actions

Android engineer

Area of Expertise

  • Information & Communications Technology

Topics

  • Android
  • App size
  • CI
  • Kotlin
  • python
  • App Performance

From Trace to Truth: Deep-Diving Android Performance with Perfetto

Perfetto is one of the most powerful performance tracing tools Android developers have — but it’s also one of the most underused. In this talk, I’ll show you how to unlock its full potential.

We’ll walk through real-world Perfetto traces and learn how to:

- Capture and view traces from release builds (via traceur, adb, or Jetpack tools)
- Identify slow startup, janky frames, and frozen UI events
- Analyze rendering pipeline delays and heavy thread workloads
- Connect traces back to real code in your app

You’ll learn how to:

- Use markers and trace sections (trace("tag")) in your app
- Visualize lifecycle timing, DI setup, image loading, and more
- Use Perfetto alongside Firebase Performance and FrameMetrics

Whether you’re chasing a mysterious slow start or optimizing Compose performance, this session will help you turn messy timelines into actionable insights.

Kotlin Performance Tracing on Android: From Coroutines to Reality

Kotlin’s coroutines and Flow APIs make asynchronous programming elegant — but when things get slow, they can also make performance opaque. What’s actually happening behind all those suspend functions and dispatchers?

In this talk, we’ll explore what Kotlin’s concurrency features really look like under the microscope of Perfetto — Android’s modern performance tracing tool. You’ll see how coroutine scopes, dispatchers, and suspending functions appear on the system trace, and how to interpret their behavior in real-world applications.

We’ll dive into how Kotlin constructs translate to threads, tasks, and context switches at runtime, revealing common bottlenecks like dispatcher overloads or nested coroutine hierarchies. Along the way, we’ll discuss techniques to instrument your own traces, correlate them with Kotlin runtime states, and visualize structured concurrency in action.

By the end, you’ll not only understand what your Kotlin code is really doing on the CPU — you’ll have practical strategies for profiling, debugging, and optimizing coroutines in production Android apps.

KotlinConf 2026 Sessionize Event Upcoming

May 2026 Munich, Germany

Akniyet Arysbayev

Delivery Hero (Berlin), Senior software engineer at App Performance team

Berlin, Germany

Actions

Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.

Jump to top