Tag Archives: #MachineLearning

Taking the leap to pursue a passion in Machine Learning with Leigh Johnson #IamaGDE

Welcome to #IamaGDE - a series of spotlights presenting Google Developer Experts (GDEs) from across the globe. Discover their stories, passions, and highlights of their community work.

Leigh Johnson turned her childhood love of Geocities and Neopets into a web development career, and then trained her focus on Machine Learning. Now, she’s a staff software engineer at Slack, a Google Developer Expert in Web and Machine Learning, and founder of Print Nanny, an automated failure detection system and monitoring system for 3D printers.

Meet Leigh Johnson, Google Developer Expert in Web and Machine Learning.

Image shows GDE Leigh Johnson, smiling at the camera and holding a circuit board of some kind

GDE Leigh Johnson

The early days

Leigh Johnson grew up in the Bronx, NY, and got an early start in web development when she became captivated by Geocities and Neopets in elementary school.

“I loved the power of being able to put something online that other people could see, using just HTML and CSS,” she says.

She started college and studied Latin, but it wasn’t the right fit for her, so she dropped out and launched her own business building WordPress sites for small businesses, like local restaurants putting their menus online for the first time or taking orders through a form.

“I was 18, running around a data center trying to rack servers and teaching myself DNS to serve my customer base, which was small business owners,” she says. “I ran my business for five years, until companies like Squarespace and Wix started to edge me out of the market a little bit.”

Leigh went on to chase her dream of working in the video game industry, where she got exposed to low-level C++ programming, graphics engines, and basic statistics, which led her to machine learning.

Image shows GDE Leigh Johnson, smiling at the camera and standing in front of a presentation screen at SFPython

Machine learning

At the video game studio where she worked, Leigh got into Bayesian inference.

“It’s old school machine learning, where you try to predict things based on the probability of previous events,” she explains. “You look at past events and try to predict the probability of future events, and I did this for marketing offers—what’s the likelihood you’d purchase a yellow hat to match your yellow pants?”

In the first month or two of trying email offers, the company made more small dollar sales than they typically made in a year.

“I realized, this is powerful dark magic; I must learn more,” Leigh says.

She continued working for tech startups like Ansible, which was acquired by Red Hat, and Dave.com, doing heavy data lifting.

“Everything about machine learning is powered by being able to manipulate and get data from point A to point B,” she says.

Today, Leigh works on machine learning and infrastructure at Slack and is a Google Developer Expert in machine learning. She also has a side project she runs: Print Nanny.

Image shows circuit board with fan next to image of its schematics

Print Nanny: Monitoring 3D printers

When Leigh got into 3D printing as a hobby during the COVID-19 shutdown, she discovered that 3D printers can be unreliable and lack sophisticated monitoring programs.

“When I assembled my 3D printer myself, I realized that over time, the calibration is going to change,” she says. “It's a very finicky process, and it didn't necessarily guarantee the quality of these traditional large batch manufacturing processes.”

She installed a nanny cam to watch her 3D printer and researched solutions, knowing from her machine learning experience that because 3D printers build a print up layer by layer, there’s no one point of failure—failure happens layer by layer, over time. So she wrote that algorithm.

“I saw an opportunity to take some of the traditional machine intelligence strategies used by large manufacturers to ensure there’s a certain consistency and quality to the things they produce, and I made Print Nanny,” she says. “It uses a Raspberry Pi, a credit card-sized computer that costs $30. You can stick a computer vision model on one and do offline inference, which are basically predictions about what the camera sees. You can make predictions about whether a print will fail, help score calculations, and attenuate the print.”

Leigh used Google Cloud Platform AutoML Vision, Google Cloud Platform IoT Core, TensorFlow Model Garden, and TensorFlow.js to build Print Nanny. Using GCP credits provided by Google, she improved and developed Print Nanny with TensorFlow and Google Cloud Platform products.

When Print Nanny detects that a print is failing, the user receives a notification and can remotely pause or stop the printer.

“Print Nanny is an automated failure detection system and monitoring system for 3D printers, which uses computer vision to detect defects and alert you to potential quality or safety hazards,” Leigh says.

Leigh has hired team members who are interested in machine learning to help her with the technical aspects of Print Nanny. Print Nanny currently has 2100 users signed up for a closed beta, with 200 people actively using the beta version. Of that group, 80% are hobbyists and 20% are small business owners. Print Nanny is 100% open source.

Image shows a collection of 3D-Printed parts

Becoming a GDE

Leigh got involved with the GDE program about four years ago, when she began putting machine learning models on Raspberry Pis and building robots. She began writing tutorials about what she was learning.

“The things I was doing were quite hard: TensorFlow Light, the mobile device of TensorFlow—there was a missing documentation opportunity there, and my target platform, the Raspberry Pi, is a hobbyist platform, so there was a little bit of missing documentation there,” Leigh says. “For a hobbyist who wanted to pick up a Raspberry Pi and do a computer vision project for the first time, there was a missing tutorial there, so I started writing about what I was doing, and the response was tremendous.”

Leigh’s work caught the eye of Google staff research engineer Pete Warden, the technical Lead of the TensorFlow Mobile team, who encouraged her, and she leveraged the GDE program to connect to Google experts on TensorFlow and machine learning. Google provides a machine learning course for developers and supports TensorFlow, in addition to its many AI products.

“I had no knowledge of graph programming or what it meant to adapt the low-level kernel operations that would run on a Raspberry Pi, or compiling software, and I learned all that through the GDE program,” Leigh says. “This program changed my life.”

Image shows 1 man and three women smiling at the camera. Leigh is taking the photo selfie-style

Leigh’s favorite part of the GDE program is going to events like TensorFlow World, which she last attended in 2019, and GDE summits. She hadn’t travelled internationally until she was in her 20’s, so the GDE program has connected her to the international community.

“It’s been life-changing,” she says. “I never would have had access to that many perspectives. It’s changed the way I view the world, my life, and myself. It’s very powerful.”

Leigh smiles at the camera in front of a sign that reads TensorFlow for mobile and edge devices

Leigh’s advice to future developers

Leigh recommends that people find the best environment for themselves and adopt a growth mindset.

“The best advice that I can give is to find your motivation and find the environment where you can be successful,” she says. “Surround yourself with people who are lifelong learners. When you cultivate an environment of learning around you, it's this positive, self-perpetuating process.”

Cloud AutoML: Making AI accessible to every business

https://img.youtube.com/vi/GbLQE2C181U/maxresdefault.jpg Jia Li Head of R&D ”Cloud AI”
When we both joined Google Cloud just over a year ago, we embarked on a mission to democratize AI. Our goal was to lower the barrier of entry and make AI available to the largest possible community of developers, researchers and businesses.


Our Google Cloud AI team has been making good progress towards this goal. In 2017, we introduced Google Cloud Machine Learning Engine, to help developers with machine learning expertise easily build ML models that work on any type of data, of any size. We showed how modern machine learning services, i.e., APIs—including Vision, Speech, NLP, Translation and Dialogflow—could be built upon pre-trained models to bring unmatched scale and speed to business applications. Kaggle, our community of data scientists and ML researchers, has grown to more than 1 million members. And today, more than 10,000 businesses are using Google Cloud AI services, including companies like Box, Rolls Royce Marine, Kewpie, and Ocado.


But there’s much more we can do. Currently, only a handful of businesses in the world have access to the talent and budgets needed to fully appreciate the advancements of ML and AI. There’s a very limited number of people that can create advanced machine learning models. And if you’re one of the companies that has access to ML/AI engineers, you still have to manage the time-intensive and complicated process of building your own custom ML model. While Google has offered pre-trained machine learning models via APIs that perform specific tasks, there's still a long road ahead if we want to bring AI to everyone.


To close this gap, and to make AI accessible to every business, we’re introducing Cloud AutoML. Cloud AutoML helps businesses with limited ML expertise start building their own high-quality custom models by using advanced techniques like learning2learn and transfer learning from Google. We believe Cloud AutoML will make AI experts even more productive, advance new fields in AI, and help less-skilled engineers build powerful AI systems they previously only dreamed of.


Our first Cloud AutoML release will be Cloud AutoML Vision, a service that makes it faster and easier to create custom ML models for image recognition. Its drag-and-drop interface lets you easily upload images, train and manage models, and then deploy those trained models directly on Google Cloud. Early results using Cloud AutoML Vision to classify popular public datasets like ImageNet and CIFAR have shown more accurate results with fewer misclassifications than generic ML APIs.


Here’s a little more on what Cloud AutoML Vision has to offer:
  • Increased accuracy: Cloud AutoML Vision is built on Google’s leading image recognition approaches, including transfer learning and neural architecture search technologies. This means you’ll get a more accurate model even if your business has limited machine learning expertise.
  • Faster turnaround time to production-ready models: With Cloud AutoML, you can create a simple model in minutes to pilot your AI-enabled application, or build out a full, production-ready model in as little as a day.
  • Easy to use: AutoML Vision provides a simple graphical user interface that lets you specify data, then turns that data into a high quality model customized for your specific needs.



    “Urban Outfitters is constantly looking for new ways to enhance our customers’ shopping experience," says Alan Rosenwinkel, Data Scientist at URBN. "Creating and maintaining a comprehensive set of product attributes is critical to providing our customers relevant product recommendations, accurate search results, and helpful product filters; however, manually creating product attributes is arduous and time-consuming. To address this, our team has been evaluating Cloud AutoML to automate the product attribution process by recognizing nuanced product characteristics like patterns and necklines styles. Cloud AutoML has great promise to help our customers with better discovery, recommendation, and search experiences."


    Mike White, CTO and SVP, for Disney Consumer Products and Interactive Media, says: “Cloud AutoML’s technology is helping us build vision models to annotate our products with Disney characters, product categories, and colors. These annotations are being integrated into our search engine to enhance the impact on Guest experience through more relevant search results, expedited discovery, and product recommendations on shopDisney.”

    And Sophie Maxwell, Conservation Technology Lead at the Zoological Society of London, tells us: "ZSL is an international conservation charity devoted to the worldwide conservation of animals and their habitats. A key requirement to deliver on this mission is to track wildlife populations to learn more about their distribution and better understand the impact humans are having on these species. In order to achieve this, ZSL has deployed a series of camera traps in the wild that take pictures of passing animals when triggered by heat or motion. The millions of images captured by these devices are then manually analysed and annotated and with the relevant species such as elephants, lions, and giraffes, etc., which is a labour-intensive and expensive process. ZSL’s dedicated Conservation Technology Unit has been collaborating closely with Google’s CloudML team to help shape the development of this exciting technology, which ZSL aims to use to automate the tagging of these images—cutting costs, enabling wider-scale deployments, and gaining a deeper understanding of how to conserve the world’s wildlife effectively."


    If you’re interested in trying out AutoML Vision, you can request access via this form.

    AutoML Vision is the result of our close collaboration with Google Brain and other Google AI teams, and is the first of several Cloud AutoML products in development. While we’re still at the beginning of our journey to make AI more accessible, we’ve been deeply inspired by what our 10,000+ customers using Cloud AI products have been able to achieve. We hope the release of Cloud AutoML will help even more businesses discover what’s possible through AI.

    By Jia Li, Head of R&D, Cloud AI, and Fei-Fei Li, Chief Scientist, Cloud AI