Skip main navigation

Case study: Victoria Lestari

We spoke to Victoria Lestari, a software engineer at Google. Victoria spoke to us about her experiences of working in the machine learning industry.
2.4
Hi, my name is Victoria. I’m a software engineer at Google, and I’ve been working for two years here. I’m working at Google on the Android machine learning team. As the name says, it means that we are building machine learning features on Android. My role is to support the machine learning platform in the infrastructure side. Since I’m working on a mobile platform, my job is to make sure machine learning is private and secure and doesn’t consume too much resources such as memory, battery, or storage on device. My typical day is a combination of checking emails in the morning, fixing bugs, working on new projects, and communicating with the other teams as well.
52.9
My pathway into working in the machine learning industry is quite generic, actually. I was majoring in computer science. I did my Bachelor and Masters, both in computer science. I came into computer science from a computer science background, but that doesn’t mean I didn’t struggle because going into uni, I didn’t know any coding. I studied really hard, especially about data science, data structures, and algorithms. And then in my third year, I started to study machine learning, which is basically similar to statistics. And I really like it. I really like how versatile it is. The great thing about machine learning is that it is a very interdisciplinary field.
98.6
So whatever you like, and if you have strong fundamentals, there is no such thing as being too late getting into machine learning.
112.8
My favourite project is ensuring the quality of our app before releases. We have to do an extensive quality assurance before releasing. My project was to set up an experiment, which compares the metrics of the older released version with the new one we have on release. Because we are in the infra team, we have to make sure that updates from our features do not impact the device’s memory, battery, and storage consumption. We use some kind of A/B testing, release old and new versions to limit the populations, and then compare the performance metrics of the versions. It’s important to know that machine learning development needs to think of the user’s practicality as well.
160.3
So I would actually encourage machine learning students to take interdisciplinary courses. For example, understanding the demography of a city or country will help us understand more about data of the population. And then we can also apply this to other as well, such as medicine or biology or everything else. Of course, finally, it’s important to update ourselves about diversity and inclusion. At Google, it’s mandatory for us to receive training about diversity and inclusion every year so that we keep being reminded about it. Yes, I believe that the diversity of the team actually makes the model better because as of now, machine learning is a dominantly male white Asian field. This is very much potentially bias.
208.1
A homogeneous group cannot think of experiences that they never have. So same in the machine learning field, we need to include more people and hear from a more diverse team to detect biases and correct it. The world is not just a white Asian male population, so we need to make sure that a good machine learning model includes everyone and helps everyone.
236.9
It’s really unimaginable how machine learning plays a part in our daily lives. And it has been very encompassing. Smart assistants, keyboard suggestions, apps, chat bots in customer service, grocery suggestions, all of these are powered by machine learning. And we don’t even realise it at times. I believe that knowledge is power. So understanding machine learning will make us be more critical of how the system works and how it is developed. The biggest impact of machine learning over the next 10 years– I think as a machine learning enthusiast, I am delighted that most of the repetitive daily tasks will be automated by machines so that we humans have more space to do more creative work that machines cannot do.
284.7
I’m from Indonesia where the industrial sector is still full of people working menial tasks. We need to make sure that these people are not left out by the progression of machine learning. I have volunteered as an instructor in a Google-led curriculum in Indonesia where they work with the Ministry of Education and then three or four other giant tech startups in Indonesia. It felt great to contribute back to my society, as well as keep being updated with the academic world.

We spoke to Victoria Lestari, a software engineer at Google. Victoria spoke to us about her experiences of working in the machine learning industry. She gives us her advice on how to get started in the industry as well as sharing her thoughts about the future of machine learning and AI.

Describe your role in the machine learning industry? What does a typical day look like for you?

I’m working at Google on the Android Machine Learning team, and as the name says, it means that we are building machine-learning features on Android.

My role is to support the machine learning platform in the infrastructure side. Since I’m working on a mobile platform, my job is to make sure machine-learning is private and secure and doesn’t consume too much resources such as memory, battery, or storage on device.

My typical day is a combination of checking emails in the morning, maintenance, working on new projects and communicating with the other teams.

What was your path into the Machine Learning industry?

I was majoring in computer science. I did my Bachelor and Masters degrees, both in computer science. I came into computer science from this background, but that doesn’t mean I didn’t struggle, because going into university, I didn’t know any coding. I studied really hard, especially about data science, data structures and algorithms.

In my third year, I started to study machine learning, which is similar to statistics. I really liked how versatile it seemed. The great thing about machine learning is that it is a very interdisciplinary field. If you have strong fundamentals, there is no such thing as being too late getting into machine learning.

What do you enjoy most about your work?

Ensuring the quality of our app before releases. We have to do an extensive quality assurance before releasing. My project was to set up as an experiment, which compares the metrics of the older-released version with the new one. We have to make sure that updates from our features do not impact the device’s memory, battery, and storage consumption.

We use A/B testing to compare the performance metrics of the old and new versions. Machine learning development needs to think around the user’s practicality as well.

What advice would you give to students who want to study Machine Learning?

I would encourage machine learning students to take interdisciplinary courses. For example, understanding the demography of a city or country will help us understand more about data of the population. We can also apply this to other areas as well, such as medical medicine or biology or everything else.

Finally, it’s important to update ourselves about diversity and inclusion. At Google, it’s mandatory for us to receive training about diversity and inclusion every year so that we keep keep this in mind in our work. I believe that the diversity of the team makes the model better because as of now, machine learning is dominantly male white and Asian. A homogeneous group cannot think of experiences that they never have, so this leads to biased thinking. For example, I believe that male policy makers shouldn’t make decisions about breastfeeding because they never experience it themselves. It’s the same in the machine learning field, we need to include more people and hear from a more diverse team to detect biases and correct it.

The world is made up of a diverse mix of people and cultures, so we need to make sure that a good machine learning model includes and helps everyone.

Why is it important to learn about Machine Learning and AI?

It’s really unimaginable how machine learning plays a part in our daily lives. Smart assistants, keyboard suggestions, apps, chatbots in customer service, grocery suggestions, all of these are powered by machine-learning. We don’t even realise it at times.

I believe that knowledge is power. Understanding machine learning will make us be more critical of how the system works and how it is developed.

As a machine learning enthusiast, one of the passions I have is around how many repetitive daily tasks can be automated, so that we humans have more space to do more creative work that machines cannot do.

I’m from Indonesia where the industrial sector is still full of people working on menial tasks. We need to make sure that these people are not left out by the progression of machine learning. I have volunteered as an instructor in a Google-led curriculum in Indonesia where they work with the Ministry of Education and then three or four of other large tech startups in Indonesia. And it felt great to contribute back to my society, as well as keep being updated with the academic world.

This article is from the free online

Introduction to Machine Learning and AI

Created by
FutureLearn - Learning For Life

Our purpose is to transform access to education.

We offer a diverse selection of courses from leading universities and cultural institutions from around the world. These are delivered one step at a time, and are accessible on mobile, tablet and desktop, so you can fit learning around your life.

We believe learning should be an enjoyable, social experience, so our courses offer the opportunity to discuss what you’re learning with others as you go, helping you make fresh discoveries and form new ideas.
You can unlock new opportunities with unlimited access to hundreds of online short courses for a year by subscribing to our Unlimited package. Build your knowledge with top universities and organisations.

Learn more about how FutureLearn is transforming access to education