Registration
30 min
開場 Keynote
50 min
Overview of Japanese software developers culture and situation
I'm software engineer and community organizer (GDG Tokyo & Women Techmakers Tokyo) in Japan. I want to share Japanese culture and situation. If you have some worry, perhaps it already resolve it in Japan.
40 min
Shoko Okochi
在導入 Flutter 之前,你應該知道什麼?
Flutter 是一套在這兩年逐漸普及的 App 開發框架。然而,即使採用了 Flutter 技術,最後你或你的企業,還是在做一套 iOS/Android App,也有不可迴避的原生部分。在這場 Talk 中,我會提到建議你如何評估 Flutter 這門技術,以及怎樣避免一些在導入 Flutter 的過程中可能會遇到的問題。
40 min
Weizhong Yang
Annotation Processing
Android development is getting much "modern" compare to the early day.
There're ton's of tools we can make our code much more clean and precise.
Annotation processing is one of the powerful weapons we can use.
Dagger and many other amazing libraries are leverage on it to get rid of boilerplate used to be.
In this session we'll go through how annotation processing works and also make some live demo.
Don't be hesitate to join this session.
40 min
Tim Lin
Google Assistant (AoG)的商業實作
1. 面對語音互動的未來,企業如何定位此全新的數位通路
2. Omni-Channel的思維中,語音又如何和其他通路相輔相成的合作
3. 在技術開發之前,最重要的事
4. 案例分享
40 min
David Chi
從機器學習平台的發展觀察如何打造資料科學團隊
資料分析與機器學習需要大量的假設、實驗、驗證,如同在實驗室一般是個冗長且由少數專家一手掌握的過程。如同軟體開發發展出 DevOps,資料科學藉此成為能夠讓更多角色參與、讓更多想法能夠被測試,實踐 end-to-end 生產線的流程。除了持續交付數據及數據產品,如何建立一個可以重複進行訓練以及測試的環境與架構,是現在建置資料科學團隊時重點探討的議題。
40 min
Summit Suen
網站壓力測試透過 JMeter on GKE
ab/siege 壓測工具只能對單一個 URL 進行壓測,若要模擬大量使用者進行特定操作,如登入會員、加入購物車等會無法模擬。
本議程將介紹如何透過瀏覽器外掛程式搭配 JMeter 進行錄製,再搭配 GKE 依照測試量級快速調整 JMeter 分散式架構。
40 min
Brent Chang
Building Android Apps for 2020
How to build high quality and high compatibility Android applications.
40 min
Fung Lam
On-Device Machine Learning with TensorFlow
Come learn about how on-device machine learning can be done with TensorFlow - whether in JS, Android, iOS, or on microcontrollers
40 min
Mike Liang
Fun with Apk analyzers and shrinking Apps
The talk will consist of two parts:
The first half would be a brief introduction to what you can do with Apk analyzer — look at merged manifest, shrunk dex files, resources, and even diffing between apks.
The second half would cover how one can shrink their apps: Proguard, R8, vector drawable and other resource optimization, aab...etc
議程將分為上下兩半場:
上半場——對 Apk analyzer 的簡單介紹,包括如何透過 apk analyzer 查看合併後的 manifest、dex、resources,甚至是比較兩個不同版本的 app。
下半場——介紹縮小 apk size 的技巧,例如 Proguard、R8、vector drawable、aab。
40 min
Teng-pao Yu
How to use AutoML Vision Python API
Use Python to call AutoML.
40 min
Sunny Au Yeung
自然語言如何做遷移學習-BERT
分享自然語言理解領域裡面,要如何使用預先訓練模型來進行語意分類與理解,內容包含BERT基本概念介紹外,也分享關於BERT的幾個應用以及其他類型的預訓練概念的模型,協助大家了解最近自然語言理解(NLU)熱門的主題火熱的原因。
40 min
Jerry Wu
Build go kit Microservices at Kubernetes with ease
快速在 Kubernetes 上搭建 go-kit 微服務應用程式,並整合 hashicorp consul, zipkin, Prometheus 等服務
40 min
Kai-Chu Chung
Lunch
60 min
捏窩-讓每個人都能成為自己生活中最有品味的創作者
談談為什麼我們要舉辦新型態的聚會活動。讓藝術人文成為一種IT技能,開始新的文化技術世代。
15 min
Chia Chiang Li
Android Fireside Chat
Android 的開發現況與中午閒談
30 min
Takuma Lee
彥彬 洪
min
Exploring AI’s Creative Potential: Generative Art & AI
Introduce generative art and how humans can rethink and explore AI's potential through artists' perspectives and artworks. The talk will start with defining generative art and then raise the common questions regarding the application of AI in art by navigating generative artworks that use machine learning.
15 min
Yu-Ying Lee
min
min
Productive web development, powered by AMP
AMP 是由 Google 領導的開源網頁框架,讓開發者能快速打造出高效能且具備良好使用者體驗的的產品。本議程概括 AMP 發展歷程與特色,幫助新手快速理解並上手,同時提供有經驗者最新的產品更新。
40 min
Irene Chang
從零開始 Flutter 開發生活: 在 Beta 時撰寫 Player Plugin 是否搞錯了什麼?
這個 Session 無法教會你寫 Flutter、更無法幫助你快速寫出一個 Player Plugin...
「純粹開發經驗、解決問題的脈絡的分享」
或許從中能得到的,只有追求卓越的態度、以及身為工程師對於架構的執著!
40 min
Hsuanhao Wang
Kotlin黑魔法:Delegated Properties
- 什麼是Delegated Properties
- 如何使用Kotlin的Delegate:Lazy, Observable, Map, NotNull
- 自己做自己Delegated Properties
40 min
Hsieh EnPing
開發Action on Google的正確姿勢[入門]
從他人上架的Action on Google中來學習該如何開發,並簡單介紹dialogflow與上架與測試。
40 min
Martin Hsu
10 years of Go
Golang 從 2009 年公開之後到 2019 也要十年了。在這場演講之中,講者會介紹十年來 Golang 的演進,也介紹為什麼你應該要使用 Golang 來開發你的系統。
40 min
Evan Lin
the Rising
AMP is so amazing and convenience for developers and users. That's why we decide to use AMP as our mobile pages in 2019. After we amplify main pages with AMP. To build a PWA with AMP become our main purpose. I would like to introduce how TW Yahoo! marketplace's work. TW Yahoo! marketplace's PWA has full functions as same as the exist mobile pages. Users could easy find and buy products through our AMP pages. Besides that it also come with vivid interactions with amp-components and smooth flow for our users. We bring wonderful user experience and amazing performance together. Let our user has home feeling when visit our PWA. Ready for it ? Let's take a tutorial about how TW Yahoo marketplace build PWAmp.
40 min
Paul Li
Beginning RxJava in Android
- How to start with RxJava in Android (sample code use Kotlin)
- How does Observable work
- How to use RxJava and Retrofit to call API
- 如何在 Android 內開始使用 RxJava。(範例會用Kotlin)
- 介紹 Observable 相關運作流程。
- 使用 RxJava 與 Retrofit。
40 min
Kelly Hong
Exploring Jetpack Compose
Jetpack Compose is a new way of building the UI on Android. As usual, if Google said you can still use existing code, it doesn't mean you don't have to learn new one. You should.
40 min
Kami Chen
Google Play Instant: Improving User Engagement
By Google Play Instant, it allow user use your app without download. The idea is like eating before buying in the market. In Viki, we use this technology to promote user engagement and increase watch time.
40 min
Yu Chuan Chang
FunWater => 整合自動化建構 + 無伺服器應用在 Action on Google
介紹開源專案 FunWater(找水喝喝),如何將台北市的 OpenData 轉換為 Action on Google 的應用。
1. 介紹專案緣起
2. 介紹 AoG
3. 介紹 Dialogflow
4. 介紹 Cloud source Repository
5. 介紹 Cloud functions
6. 介紹 Algolia,利用 GEO search,找出離自身最近的飲水機
7. 介紹 Google Maps API 使用 distance_matrix 取得點與點所需時間
8. 介紹 Cloud build ,並設定環境變數
9. 介紹 Cloud schedule 設定排程自動更新 OpenData 資料
40 min
WeiHao Huang
Awesome web data visualizations with Google Maps Platform and deck.gl
The Google team recently announced initial support for the Deck.gl framework, which brings beautiful, WebGL-accelerated visualization of large data sets to Google Maps. In this talk we’ll look at how to integrate Deck.gl into maps on the web using the Maps JS API.
40 min
Alex Muramoto
Hello Cloud Run
Cloud Run 是 Google Cloud 新推出的 Serverless 服務,可以很容易地部署無狀態網路應用容器。由於實作採用了 Knative 專案,使用者也可以自行於 Kubernetes 叢集上建立相同的環境運行應用容器。
本次講者將為各位介紹 Cloud Run 的特色、與其他 Google Cloud 服務的整合,以及應用案例。
40 min
Bear Su
Coffee Break
20 min
2019 Women Techmakers scholars 經驗分享
share Google WTM scholarship
40 min
Aliyah Hsu
Chia-Chi Hsu
Mia Yu
芝蓶 甯
Introduction to the CameraX Jetpack Library
Learn how to get started with CameraX support library in Jetpack, which provides the best support across a wide selection of Android Camera hardware with minimal effort, and learn from teams at Google that have been using CameraX in their apps.
40 min
Max LY Lau
TY Chang
Kotlin Coding dojo
使用語言 kotlin
活動內容
1. pair programming 形式進行,一組 7 分鐘
2. 一位 driver (輸入程式碼的人),一位 navigator (審查程式碼的人)
下一輪時 navigator 變成 driver 跟下一個 navigator 再變成一組
3. 採自願排隊上台輪流 pair
活動規則
1. 要寫測試
2. 有紅燈時要先修測試
3. 有紅燈時不能重構
4. 要寫 production code 要有測試
5. 每一輪時間到的時候至少要可以編譯
85 min
Andy Yang
Cuber Yuan
Google assistant development solution
Google assistant 實作方法 (AoG 、App action、Smart home)
40 min
KEVIN chiu
用BigQuery帶你快速完成第一個ML專案
第一次接觸BigQuery就讚嘆天下竟然有這麼神奇好用的東西。
本演講難度入門,適合沒有經驗,會寫一些 SQL 語法,想要試試機器學習模型的人,又或者是,對於資料科學家來說,BigQuery佐Cloud Dataprep佐Data Studio簡直是找到知音!演講內容包含如下:
1. 建立資料集,使用 Cloud Dataprep UI 做資料準備、清理
2. 使用 Data Studio 視覺化建立 dashboard
3. 引入 ML 到 BigQuery
40 min
Rowan Tseng
12 Things to Know When You Run Your App on GKE
Kubernetes is a de facto standard container orchestrator. As well as best practices of Docker and Kubernetes, you should know those of GKE. GKE is a fully managed version of Kubernetes but it’s optimized for GCP. So to make your application running on GKE most efficient and utilized, you need better understanding of GCP. I will cover 12 major things to care about GKE. I will also share the latest update of Kubenetes and GKE.
40 min
Jun Sakata
以 Ktor 快速打造 Web 應用
Kotlin 是一個適用於多平台的程式語言。為了讓開發者更輕鬆打造 Web 應用,JetBrains 研發了 Ktor Web 框架及 Exposed SQL 函式庫,讓大家可以用精簡清楚的語法,快速打造 Web 應用。在這場分享裡,將會以實作一個簡易的 CRUD 應用程式為例,展示 Kotlin 在 Web 方面的實務應用。
40 min
范 聖佑
Enjoy DI life with Koin
When mentioning about DI framework on Android, in addition to Dagger2, there is actually another alternative, Koin. Koin is a lightweight dependency injection framework written in pure Kotlin. As a DSL, Koin provides a pragmatic API that significantly simplifies the injection process.
This session will talk about how easy to set up the dependencies in Koin and how easy to inject instances. Besides, We will demonstrate how to integrate Koin with Model-View-Presenter (MVP) and Model-View-ViewModel (MVVM). Lastly, we will talk about testing with Koin.
40 min
Aaron Chu
min
Google Cloud Run used in chatbot/action webhook
how to use Google Cloud run in chatbot/action webhook.
40 min
Wolke Lin
Getting started with microcontrollers and Google Cloud IoT Core
In this talk, we will start from a famous and affordable microcontroller unit. We will first discover how to write code for this MCU so we can get environment data. Then we will connect the microcontroller to the Cloud using Google Cloud IoT Core. Our goal is to discover what these online Cloud IoT solutions offer, and why you may (or may not) need those.
We will focus on Google Cloud IoT core, core concepts remaining the same on other online platforms.
40 min
Gautier Mechling
Deploy your Application from Container to Kubernetes
回顧 .jar 和 .war 的年代到現在,部屬的結構、技巧與精神其實沒有改變多少。 但軟體世界總是快速的更迭著,這些年無論是在手法、量級和規範等等都大幅的改變我們部屬專案的樣貌。
我們可能不再只是載了一個 Tomcat 然後把 .war 丟上去 VM 開起來,然後再擋個反向代理直接做分流。 這些年 .jar 和 .war 連同環境被容器化了!我們從專案包起來跑在 JVM 上,演變成專案連同執行環境包起來跑在 Docker 上。 維度上我們跳躍了一個層級之後,現在因應容器的管理議題,容器編排工具戰爭目前看起來暫告一個段落,K8s (Kubernetes) 脫穎而出。
在這個議程裡,我想跟大家分享一些心得:
1. 以前我們怎麼部屬的?隨著量級增長我們怎麼處理常見的議題?
2. 從容器化到 K8s 的改變及影響,然後 Demo。
3. 回顧過往,展望未來,做個總結。
#Deployment #Docker #Kubernetes #GKE
40 min