How did you join Scale?
Both Alex, Scale's founder and CEO, and I went to MIT and frequently played poker together on MIT's poker club. When Scale went through YC, I was also in Mountain View working as a ML engineer at a large company. I watched as Scale found product-market fit at a lightning-fast speed. I've always been very interested in AI/ML, so once I saw explosive growth in a friend's company in that space, I knew I had to experience it firsthand. I interned with Scale in the following months, and I joined full-time shortly after that.
What about the company’s mission is the most compelling to you?
Scale's mission is to accelerate the development of AI applications. In the short term, Scale is focused on providing large quantities of high-quality ground truth data to companies using cutting-edge deep learning techniques. We started out working mostly with companies in the AV space, and as companies in other computer vision verticals — like robotics, retail, and AR/VR — grow, we have begun working with them as well. What these companies are accomplishing is incredible and the opportunity to help them scale is both interesting from a technical standpoint as well as immensely satisfying from an impact perspective.
With a vast network of labellers doing Scale tasks around the world and an optimized task pipeline serving them tasks, we can now explore providing more diverse types of ground truth data as well. However, we aren't just a data labelling company. We want to help machine learning engineers in the most useful ways that we can; right now we believe that is data labelling, but our platform could be extended to many parts of the AI application development stack. The possibilities in this space are endless with our incredible team, and I'm excited to work with our customers and labellers towards these goals!
You worked at big companies prior to joining Scale. What was the transition like? What advice do you share with others contemplating a similar move?
At a large company, you often have the luxuries of product development cycles that take months, well-defined problems, and structured mentorship programs. At a startup, it might be optimal to develop a completely new product or feature to test out a hypothesis in a week — and the requirements and goals of that product are the responsibility of you and a very small team! Everyone is so busy at a startup that there's not a lot of time for explicit mentorship, but there are countless chances to test your skills and assumptions with a trial by fire.
I find this sort of work environment very motivating, and I've learned a ton about growing a company — not only as a product manager, but as an stand-in engineer, customer success representative, designer, QA-er, recruiter, data analyst, marketer, operator, and more— in my time at Scale. Also, with a rapidly growing team, I'm learning a lot from other Scale team members about how to best scale product development. For others considering a similar move, I'd recommend thinking deeply about your goals and your desired day-to-day. If you don't mind trading structure and certainty for the personal growth and fulfillment that can accompany working at a hyper-growth startup, it might be a good move!
What is the most exciting project you’ve worked on?
When I first joined Scale, we just offered 2D annotation products. A few of our self-driving customers mentioned off-hand that they would also potentially use a 3D annotation product. After roughly sizing the market and contemplating the trade-offs, we decided to build a minimum viable product to see if this additional business was a good direction for the company. I worked with a few engineers to build the first version of this product that could support our pilot customers. I frequently talked with our customers and taskers, watched videos of their experiences, and did tasks myself to spec out features that would hopefully make our taskers more efficient and customers more satisfied with the results.
By obsessing over the metrics of data quality and throughput and iterating quickly, we increased our total 3D business (by tasks and by customers) many-fold in just a few months. With the continued brilliant work of many other Scale team members, 3D annotation has many happy customers! Experiencing the early engineering, design, sales, and customer support that is required to go from “0 to 1” easily makes early development of the 3D annotation tool the most exciting project I've ever worked on.
What are some of your favorite things about the culture?
Scale's culture was a large contributing factor to my joining. Scale has enumerated the qualities we are looking for in future hires, and we stick to our principles even when it is tempting to compromise (e.g. to reach a hiring goal). As a result, everyone at Scale is extraordinarily passionate and determined to reach our company goals. Scale team members feel responsibility for the outcomes of the company and believe that their actions can change these outcomes. This prevailing optimism and the close-knit quality of the Scale team make even the worst days bearable and the best days insanely great. Additionally, Scale's culture of intelligent discussion and decision-making rigor has made me a much stronger product manager and more rational person.
What is the most valuable thing you’ve learned here?
The most valuable thing I've learned at Scale is to be confident and relentless in pursuit of any goal. There's even more uncertainty than I thought there would be working at a startup, but with a clear, intelligent mind and similar team, you can achieve unbelievable outcomes.