Saad Bin Amjad
Technical Lead, Monstarlab Bangladesh
Google I/O
Challenges of Present
Focus on Problem-Solvers
What's New
Bangladesh's Future
Follow up
2019
2020
May 19th 2021
Me
was a virtual sandbox for registered attendees and visitors to get a hands-on experience with Google's new products and features.
The COVID-19 pandemic has significantly changed the way the world operates.
More than 5 billion people now use the internet as of 2021.
Applications need to be shipped out fast and efficiently that ensures best user experience.
Developer experience and their ease is top priority.
Dependency
Capabilities
Integrations
Accessibility
Performance
Security
Engagement & Experience
Developer Tools & Experience
Developer Tools and Experience
- Are there right tools available?
- Are they easy to learn and use?
- How easily can I solve complex problems?
- Code once, ship everywhere?
- Faster to production.
Accessibility
- Can we create something that is accessible by many?
Dependency
- Governments, financial institutions.
- Offices, business, schools etc.
Engagement & Experience
- How to improve user's experience
- How to monitor metrics
Security
- Is User data safe and well protected?
Performance
- Under immense load, can it work and be stable?
- Does it consume too much power?
Capabilities
- Can it run on multiple devices?
- Is the experience same across devices?
- Can different hardware devices be accessed?
Integrations
- How does other 3rd party services be seamlessly
be integrated?
Maya, the digital health assistant's partnership with Google has increased the accuracy of Maya's natural language processing to above 90%. In 2020, Maya's platform reached over 10 million users.
Web
Machine Learning
Android
Firebase
Flutter
Web
Machine Learning
Android
Firebase
Flutter
In browsers you can enjoy video and audio, first-class platform with features background blur etc.
Using JS you can rotate and tilt maps in 3D.
You can access more devices like play games with gamepad from the browser, create IDEs in browser since file system can be accessed.
Most of the apps now have PWAs and runs on web and also installable, eg TikTok, Rakuten.
And PWAs are now easily highlighted and listed in the Play Stores with matured TWA apis.
Computing power to browser: Web Assembly
Evolved and Matured WebRTC
V8 is faster.
Compact image format: AVIF.
BFcache, lazy load images.
Maps JavaScript API WebGL beta!
More Hardware APIs
Core Web Vitals
Moving away from 3rd party cookies, with solution like FLoC etc
Web
Machine Learning
Android
Firebase
Flutter
ML expert
Complex Features to start solving from scratch
Different Platforms to harness from
Mobile | Web | Cloud
TensorFlowHub for existing models.
Android + TensorflowLite Task library
TensorFlowLite provides the framework for a trained TensorFlow model to be compressed and deployed to a mobile or embedded application.
ML in web: Tensorflow.js all in browser
Mobile | Web | Cloud
Vertex AI
Simplifed, experiment and train
Train machine learning models without user datasets, Online console.
Transparent: Metrics to show what models are relying on to make predictions in order to improve transparency.
Web
Machine Learning
Android
Firebase
Flutter
"Keep the openness, become opinionated about the right way to do things and make the right way the easiest way"
- Karen Ng
Audit app's data request
Performance gains: standby bucket, foreground services,
Improved startups, animations and widgets.
Android Studio Artic Fox (Beta)
- Test results in devices, debug databases
Android Jetpack: Macrobenchmark (Alpha)
Benchmark your app startup and integrated behaviours like scrolling performance.
Kotlin Coroutines
Kotlin Symbol Processing: an all-new tool for building lightweight compiler plugins in Kotlin
Jetpack Compose 1.0 in July
Modern design practices UI toolkit
Material You components, WearOS.
Google Assistant: Voice shortcuts for your app
Single smartwatch wearable platform
Developers to create apps and widgets for a single OS instead of splitting their efforts between Wear OS and Tizen.
Bringing in Samsung in the fold.
New Health APIs.
Web
Machine Learning
Android
Firebase
Flutter
Faster to production, instead of working from scratch:
Firebase Extensions, pre packaged solution.
Extensions Stripe, Algolia, MailChip etc.
Firebase Remote Config
Without releasing versions
You can test features behaviours from groups of user
Personalisation: For example, you can tweak the homescreen for users based on their country.
Web
Machine Learning
Android
Firebase
Flutter
Open Source
Beautiful UI experience across multiple platforms
Compile code into machine code for:
Android, iOS, Web, Desktop etc..
Productivity increased, ship app and features faster.
Reduction in Codebase: Google Pay unified its code.
Reduced code by 45%.
Flutter 2.2 announced
Improved Beta Support for Flutter on Desktop
Null Safety
Dev Tools are added, e.g to look at memory usage
Google Pay payment Plug-in
Joining Forces with Firebase, Material You
Bangladesh's Future
Developed By:
Twitter: https://twitter.com/saadbinamjad
Github: https://github.com/saad-amjad
LinkedIn: https://www.linkedin.com/in/saadbinamjad