You hear a lot about artificial intelligence these days, but what you don’t hear much about is a small team that’s actually dreaming up and implementing it for big-name clients. Strong Analytics has only been around a couple of years, but they’ve already created a name for themselves as the sort of team that can bring big, state-of-the-art technology-driven projects to fruition. We stopped by to see the mighty little team in action.
FOUNDED: 2016
EMPLOYEES: 11
WHAT THEY DO: Strong helps companies design, build and deploy custom machine learning and artificial intelligence products and solutions. How cool is that?
WHERE THEY DO IT: Chicago — across the street from Marina City.
NO COMMUTE WEDNESDAYS: The whole team works from home on Wednesdays.
BIG NAME CLIENTS: Strong works on projects with companies of all different sizes and has worked with Mercedes-Benz, Dictionary.com and Sunsweet in the past.
Juan Llanos, Senior Data Scientist
Juan leads projects that use natural language processing and deep learning.
BEYOND WORK: Juan is really into video games, particularly multiplayer ones, which gives him the opportunity to keep in touch with his far-flung friends.
What’s unique about working at Strong?
One of the most unique and exciting aspects of Strong is the pace of our work. By the end of my first day, I felt equipped with the tools I needed to dive straight into the core of my first project. By the end of my first month, we had already researched, developed and deployed a machine learning application that was analyzing tens of millions of real-world texts per day. One of the things Strong does well is keep work fun and light-hearted. Whether we’re discussing Python’s new walrus operator, sharing a GIF on Slack or playing a quick Mario Kart tournament after work, you always feel a sense of camaraderie.
We hear a lot about machine learning these days, but you’re actually on the front lines. Is it the panacea we’ve heard it is?
There’s a lot of hype surrounding machine learning, but there are very few teams that can actually take a project from research to deployment. We hear a lot about the major Bay Area tech companies using state-of-the-art techniques for their products, but there’s tremendous value in applying machine learning in other industries and organizations of all sizes. You can get a sense of this from our client list, which includes companies across several industries.
One of the things Strong does well is keep work fun and light-hearted. [...] You always feel a sense of camaraderie.”
What projects have you had the opportunity to work on here?
I lead the development of a custom end-to-end research platform that uses several deep learning models to identify cyberbullying and other abusive content that children are exposed to online. I’m proud of the impact that this project has every day, as it allows millions of kids to freely explore the online world while also keeping them safe. I am also proud of the technical challenges that we’ve had to overcome, from building models that understand children’s constantly evolving vocabulary to scaling these models to the tens of millions of online text and media pieces that children are exposed to daily.
Jacob Zweig, Co-Founder
Jacob leads projects that are focused primarily on deep learning and computer vision.
BEYOND WORK: When he’s not teaching computers to see, Jacob enjoys gardening.
When Strong started, it was just another company in a crowded industry. How have you managed to differentiate yourself?
While the field of data science grows at a rapid rate, the number of teams that take a project from design to production remains limited. At Strong, we have extensive experience doing so in many industries, with deployed solutions that help keep millions of kids safe on the internet, reduce prescription fraud across the United States, monitor high voltage power lines and more. Our ability to build such effective solutions starts with a thoughtful and user-centered approach. We work with our clients to understand their business and goals before identifying high-value opportunities to apply machine learning and artificial intelligence. Armed with a deep understanding, we are able to design and implement effective tools to accelerate their business.
At Strong, we build tools that are actually deployed and interact with millions of users each day.”
What about your company inspires you? What keeps getting you excited to get out of bed and come into the office in the morning?
It’s incredibly exciting to see the impact of the work that we do. In consulting, it’s unfortunately common to deliver a deck that goes into someone’s file drawer and never sees the light of day. At Strong, we build tools that are actually deployed and interact with millions of users each day. We’ve helped fruit growers in remote Chilean regions earn better wages by creating portable deep learning-based fruit grading machines. We’ve helped optimize the distribution of life-saving blood to hospitals. The list goes on, and it’s really rewarding to see such a tangible impact on the world — it’s hard to achieve that in a lot of careers.
Strong is a tight-knit group, but you’re planning on growing the team. Who does well here?
We’re a quirky bunch with a vibe that oscillates from intellectual to whacky. Just the other day, we went from a discussion on do-calculus and causal inference to trying to figure out if the phrase ‘on fleek’ is still cool (it turns out it is not…). We have a super collaborative environment, and we look for candidates that are curious, passionate, collaborative and driven.
We maintain a fast-paced environment where we have the opportunity to work with a team of experienced data scientists to build and deploy state-of-the-art products with machine learning and artificial intelligence. This means that you should feel comfortable reading, implementing and improving upon techniques from literature, writing clean and performant code, sharing your knowledge with the team and enjoy building awesome things!
Brock Ferguson, Co-Founder
Brock leads projects focusing on data architecture, engineering and causal modeling and inference.
BEYOND WORK: When he’s not in the office, Brock likes to get away somewhere warm with his surfboard.
Starting a company is a lot of work and has a lot of risks — why did you do it?
Growing a business is definitely a different pace than academic research. At the end of the day, [my co-founder] Jacob and I both want to build things that have real-world impact, whether that’s making peoples’ jobs easier, helping businesses operate more efficiently or transforming traditional industries with modern technology. At Strong, we get to do this every day with our partners and it is still what drives us. Here, we don’t just learn about new companies and industries as outsiders — we dive into the very core of their business and help them ask the question, ‘How can we make this better?’
How do you think Strong fits into the overall, broader arena of data science?
Data science is such a broad field that it has rightfully started to fracture into subfields that, in practice, do very different things. We help companies build and ship products that leverage machine learning. Some of these products are consumer- or client-facing, while others are internal, even automating parts of our clients’ core operations. We differentiate ourselves through our team. Everyone on our team has the expertise required to take solutions from design all the way to production. It’s what we do on every project.
We don’t just learn about new companies and industries as outsiders — we dive into the very core of their business and help them ask the question, ‘How can we make this better?’”
What are some of the problems the team is currently solving?
Many of our clients are lucky to have really strong data analysts and engineers on their teams. Analysts are great at aggregating data together and using this data to make educated business decisions. Engineers are great at encoding and automating processes. They bring us in to bridge the gap between these skill sets as machine learning engineers, with our goal often being to build a product or tool that ingests data automatically, continuously learning from the data and making intelligent decisions in a completely autonomous fashion. Although we work in lots of different industries, most projects fit this same general blueprint.
