F8 2019 Day 2 keynote and session videos

Meta

At F8 2019 day 2, topics included a new method for computer vision, new techniques for self-supervised learning, and a new framework for inclusive AR/VR. Watch the video of the keynote address and …

Overview

The F8 2019 Day 2 keynote and session videos highlight advancements in AI tools, computer vision, self-supervised learning, and inclusive AR/VR frameworks. Key presentations include discussions on scaling AI experiences with PyTorch, enhancing developer productivity through machine learning, and optimizing products with adaptive experimentation.

What You'll Learn

1

How to accelerate the path from research to production deployment using PyTorch

2

Why using static analysis tools can improve code reliability

3

How to leverage machine learning for code search and automatic bug fixing

4

When to apply adaptive experimentation for product optimization

Key Questions Answered

How does Facebook use PyTorch for AI experiences?

Facebook utilizes PyTorch to transition from novel AI research to large-scale production in areas such as computer vision and natural language processing. This framework accelerates deployment and enhances the capabilities of their AI tools, making them more efficient and effective.

What tools does Facebook use to ensure reliable code at scale?

Facebook employs tools like Infer for static analysis, Sapienz for automatic test case design, and idb for iOS automation. These tools help developers identify and fix code issues early in the development process, enhancing overall code reliability.

How does Facebook enhance developer productivity with machine learning?

Facebook builds tools that utilize machine learning to identify patterns in code, which aids in code search, code recommendation, and automatic bug fixing. These tools streamline the development process and reduce the time spent on debugging.

What is the role of adaptive experimentation at Facebook?

Adaptive experimentation is an AI-enabled testing approach used by Facebook to optimize various aspects of their products, including infrastructure and marketing campaigns. This method allows for more efficient testing and better decision-making based on real-time data.

Technologies & Tools

Some links below are affiliate links. We may earn a commission if you make a purchase.

Framework

Pytorch

Used for accelerating the path from AI research to production deployment.

Key Actionable Insights

1
Integrate static analysis tools like Infer into your development workflow to catch bugs early.
By identifying issues before code is shipped, developers can save time and resources, leading to a more reliable codebase and smoother deployment processes.

2
Utilize machine learning tools for code recommendation and automatic bug fixing to enhance productivity.
These tools can significantly reduce the time developers spend on repetitive tasks, allowing them to focus on more complex problems and innovations.

3
Consider implementing adaptive experimentation to optimize your product offerings.
This approach allows for data-driven decisions that can enhance user experience and improve product performance over time.

Common Pitfalls

1

Neglecting the use of static analysis tools can lead to undetected bugs in production.

Without these tools, developers may miss critical issues that could affect user experience and system stability.

2

Failing to leverage machine learning for developer productivity may result in wasted time on repetitive tasks.

By not utilizing available tools, teams can struggle with inefficiencies that hinder their overall development speed.