rahulrav.com
Software Engineer at Google working on the Android Toolkit team.
30 posts
312 followers
113 following
Regular Contributor
Active Commenter
comment in response to
post
This is incredibly exciting ! Looking forward to your talk 🙏
comment in response to
post
Just wanted to congratulate you folks on some of the delightful animations. Also thank you for Trainline 🙂 another beautiful app.
comment in response to
post
I should really make this tool easier to discover, and use.
Being able to compare perfetto traces is fantastic. I have been working on writing something to make this part simpler as well. Hopefully will have something to share soon.
comment in response to
post
I agree. We wrote plot-benchmarks (cs.android.com/androidx/pla...) to make it simpler to read these distributions. This also makes it easy to compare benchmarks before / after normalization between runs; and also provides a very valuable metric of statistical significance.
comment in response to
post
😭
comment in response to
post
Dependency injection is a good idea but things seem to go sideways very fast in every code base that's had it (that I have touched). Maybe I am set in my ways, but I like constructors. No magic whatsoever.
comment in response to
post
Wow 😮
Looking at new UniFi gear now..
comment in response to
post
Yes ! We recently made androidx.traceprocessor KMP available. So you should be able to emit traces and use Trace processor to assert for them. That's one of the things I want to demonstrate.
comment in response to
post
That is the eventual goal. We will support a platform driver that will do exactly that. It's not there yet.
comment in response to
post
Thanks for writing this series. Been super helpful with a bunch of tests that I have been writing recently.
comment in response to
post
Performance is what gets me as well. Got to write that benchmark and tweak things 😬
comment in response to
post
Still a work in progress but getting close to being usable. Need to write benchmarks to verify that the overhead is minimal. 😅
comment in response to
post
It's the same for me. The relieved joy that sweeps over after you have accomplished something you thought was going to be hard keeps you going.
Until that happens there is almost like a background thread running trying to figure things out.
comment in response to
post
Geospatial folks refer to that as a "Null island".
comment in response to
post
I really like Kandy (kotlin.github.io/kandy/welcom...). Especially coupled with Kotlin Notebooks, it's magical.
comment in response to
post
Don't worry. It was not too bad.
I know just enough about CSS to get myself out of trouble. 😅
comment in response to
post
I will be very very careful when naming fields in protocol buffers going forward.
Also be right back after having a couple of things I defined recently 😫.
comment in response to
post
It must have been. I think the answer might be a similar API defined by the Java Microbenchmarking harness. Prior art 🙂
comment in response to
post
I don't know what the contenders were to be honest 🙂. These APIs became final before I started looking into Perf. Will ask Chris and get back.
comment in response to
post
I am glad you found this useful. Perfetto is easily one of the coolest tools that Google has built. There is also so much more stuff that I wish we had the time to show.
comment in response to
post
What do you not like about coroutines ? Genuinely curious 🤨
comment in response to
post
I also love (and perhaps a bit obsessed) with baking 🍞 and playing badminton 🏸.
comment in response to
post
Currently focused on making Android apps (& Jetpack Compose) faster. To that end, I am currently working on:
📐 Jetpack Macrobenchmarks - helps measure app performance at a macro level (startup and jank).
👤 Baseline Profiles and Dex Layout Optimizations (startup profiles) - Makes apps a lot faster.