Labelboxer Spotlight: Amanda Bell, Software Engineer

Amanda is a full stack software engineer on the Perception team at Labelbox. Her first exposure to web development was building neopets fan pages on Angelfire with her sister. Though she had planned to major in Art History at university, she switched to Computer Science when she realized how much creative thinking and actual creation took place in her engineering electives.

“As a programmer I am so lucky to be in an engineering discipline that affords fast prototyping and iteration of ideas. Taking an idea and bringing it to life is the best part of my job. Working at Labelbox I not only get to wake up and do that everyday, but I also get to work on a product that facilitates other companies to do the same, creating a chain reaction of innovation.”

— Amanda Bell, Software Engineer at Labelbox

Why did you decide to join Labelbox?

During university, I completed two software engineering internships at Google. After school, I joined an early stage cryptocurrency startup which went from a greenfield project to an app that was gaining thousands of new users a month. When I joined Labelbox, I was looking for an environment that struck a balance between an early stage startup and a big tech corp. I wanted something fast paced where I could have impact, but I also wanted room to grow my technical and skills in the way the larger corporations have more resources for.

Labelbox is a startup in its high growth phase, building a product in the constantly evolving space that is AI. You will never be bored. At the same time, there is a healthy ratio of experienced specialists to more junior engineers (we don’t hire just to get someone in the seat), reasonable but rigorous technical standards, and an emphasis on setting skill development goals each quarter in addition to the company wide product goals (along with a substantial education budget to help you out with that).

The remote work was also a big selling point for me. I can do my job from anywhere in the world with wifi, whether it’s a beach in Brazil, or slightly less glamorously, my Mom’s kitchen in rural Saskatchewan, Canada. We are given plenty of autonomy and encouraged to take advantage of it within the setting that works best for us.

What makes engineering problems at Labelbox unique?

In web development, and particularly frontend web development, it’s easy to feel like you’re solving the same problem on repeat. Even if you move to a company with a different product, or learn a new tech stack, many web applications can ultimately come down to forms and tables. Labelbox on the other hand isn’t a typical CRUD app. For instance, we get to build out drawing tools in the browser with canvas, are tasked with handling a range of different input data from geospatial tiled imagery to dog training videos, and we’re finding new ways to integrate model assisted labeling into our platform so our users can use machine learning to accelerate their labeling process. And on top of that, we need to keep it all performant and scalable. Like I mentioned above: you’ll never be bored.

What have you learned so far during your time at Labelbox?

I’ve learned so much it’s hard to zero in on one thing, but I used observables for the first time when I started working on our new video editor and it was very rewarding. As a web developer, picking up new technologies is a typical part of the job, but with RxJS I wasn’t just learning a framework with a new syntax for rendering DOM nodes (although that can be fun too). Observables required a whole new way of visualizing event streams, and once it clicked it gave me a whole new toolset for handling DOM events (or any push API for that matter). It was a huge help while working on video and made otherwise tricky tasks (like dynamically changing the width of the cell in the timeline representing a frame, or porting over our buffer detection from the old editor) into straightforward polish items.

Which Labelbox value resonates with you the most and why?

Making customers win. Not because I have a secret affinity for sales, but because it’s the most accurate way we can ensure that we are staying focused and maximizing our impact. The organizations that use Labelbox come from a range of industries, including health care, agriculture, fashion, insurance, and more. Each organization has their own unique set of problems that they’re hoping to solve using Labelbox. The more that we can continue to find the commonalities across their different issues and use cases, the more we can step in to be the general problem solvers that can accelerate current and future customers' innovations.

What are you excited about currently?

I’m excited about expanding our editors to support a wider range of data types. While we continue to maintain focus on improving our core image editor, it's also cool to see that it can be used as a foundation for additional editors that can handle video, tiled imagery, and text. From an engineering perspective, part of what made our new video editor so fun to work on was determining the best way to represent temporal labeling data. Everytime we add a new data type, we’re going to have the fascinating task of determining how to depict a new kind of knowledge in a browser window. Information comes in many different forms and the wider variety we can support, the wider variety of problems we can help our customers solve.

Want to join the Labelbox team? Explore our open positions and apply today. Stay tuned in the coming months for more spotlights on Labelbox employees!

Tags

Labelbox

Labelbox is a collaborative training data platform empowering teams to rapidly build artificial intelligence applications.