#language engineer

This month’s linguistics jobs interview is with Brent Woo, a Language Engineer for a voice assistant. Brent has an MA from Eastern Michigan University and a PhD in Linguistics from the University of Washington. In this interview, Brent describes how his career provides him with work-life balance and how his linguistics training ties into his work developing technology we use every day, but might not realize requires language data to run smoothly. 

You can find Brent online on LinkedIn. 

What did you study at university? 

Linguistics and more linguistics! I entered undergrad not really knowing what I wanted to do. I took some finance and economics with the vague idea of becoming an accountant, but I didn’t have an interest in the material. I took some political science and cultural studies classes with some interest in “international affairs” but it still wasn’t quite right for me. I took an intro linguistics class and knew immediately this was it. This was the mix between analytical thinking and interesting human issues that I was looking for. I went on to finish BAs in Linguistics and Russian Studies, and then an MA and PhD in Linguistics. What is your job? 

Language Engineers work on the language designs for voice assistants. When you say something funny like “let’s raise the roof” and the voice assistant can interpret that to raise its volume to the maximum level, that was likely due to an LE language design. LE look into common variations on requests, understand the failures in human-machine communication, and adjust recognition coverage for them. We work on the scale of millions of utterances a day, so this process also has to be automated to some degree. How does your linguistics training help you in your job? 

I’m lucky to be considered a subject matter expert in language, and can usually make compelling arguments for utterance coverage based on language and human arguments alone. A common scenario is product or business stakeholders will request to have utterance X go to feature Y, and an LE will know from their expertise that no sane human customer would utter X to trigger Y. The LE has to present the case why that wouldn’t be a good design, and propose alternatives, and overall help the feature get better language coverage. This involves classic syntax for figuring out sentence structures and variations, and semantics for the annotation schema. Sometimes phonetics comes up when there are speech recognition conflicts or misfires! What was the transition from university to work like for you? 

Graduate school was great preparation for moving into this position. I already had experience running quantitative experiments and analyzing the results, managing different stakeholders (advisor, students, lab, external collaborators), and—most importantly—dealing with difficult customers. Two things were new to me at the beginning: having conversations at the business and product level, and the pace of work. I wasn’t familiar how to navigate conversations about project budget and headcount, but I had a good friend as a mentor who helped me out with these early on. The pace of work differs from grad school: work comes and goes with the product release cycles, there are periods of crunch time and periods of relative calm and catch-up. Do you have any advice you wish someone had given to you about linguistics/careers/university? If you’re interested in exploring a non-academic path, talk to linguists in those paths, watch the panels and webinars we’ve run on these topics, ask for mock interviews, and read Superlinguo! You’ll get a sense of what background you need and what projects you should be working on. 

Any other thoughts or comments?

The most important thing for me when finding a job after graduate school was to find a work-life balance. While I’m happy and lucky to have an interesting job related to language, I was really thrilled to discover that it was very flexible and I can live a rich life outside of work responsibilities. There were other jobs that may have been a better fit, but were more demanding on my lifestyle or diplomatic capacity (for example, managing people). I have many colleagues at this and other companies who have taken on more intense positions and end up burning out or quitting due to the stress. I find the work-life balance in this position almost ideal and I am very happy.

Related interviews:

Interview with a Natural Language Annotation Lead

Interview with a Software Engineer

Interview with a Data Scientist

Interview with a User Experience (UX) Researcher

Interview with a Computational Linguist

Recent interviews:

Interview with a Data Manager & Digital Archivist

Interview with a Natural Language Annotation Lead

Interview with an EMLS/Linguistics instructor & mother of four

Interview with a Performing Artiste and Freelance Editor

Interview with a Hawaiian and Tahitian language Instructor, Translator & Radio Host

Resources:

The full Linguist Jobs Interview List

The Linguist Jobs tag for the most recent interviews

The Linguistics Jobs slide deck (overview, resources and activities)