One of Two Sigma Ventures’ core values is being “smarter together.” We believe in facing hard problems with low ego and intense curiosity about the thoughts and feedback of others around the table. Oftentimes, the best solution to a problem is nontraditional, and nontraditional solutions can be arrived at in unexpected ways.
One of the ways we have helped our portfolio companies is by running hackathons, where we tap into Two Sigma’s extensive data science and engineering talent to help solve technical problems. Two Sigma has been building a world class technical organization for 20+ years, and today calls 200+ PhDs and 1200+ data scientists and engineers employees. Being part of this organization enables us to access this large talent pool to support our portfolio companies in powerful ways, to think and work even smarter together.
We hosted Two Sigma Ventures’ first hackathon a few years ago with our portfolio company, WHOOP,* a leading consumer wearable technology company that provides a digital, 24/7 health and fitness coach. Over the course of just a few days, Two Sigma data scientists were able to help WHOOP gain insights that led to an increase in the accuracy of their machine learning (ML) algorithms that turn heart rate variability (HRV) data into insights about strain and recovery of the body, by 30%. Our take-away from this was that exciting outcomes could come to life by giving our portfolio company’s talented team access to a bench of experts who are willing to roll up their sleeves and approach problems in new ways.
This time, we partnered with our portfolio company Nelo,* a fast-growing Mexico City based fintech company, who we believe has the potential to emerge as the leading consumer financing provider in LatAm, starting with Nelo’s elegant buy-now-pay-later (BNPL) app and merchant checkout experience. We led their series A in late 2021, alongside Homebrew and Susa. The hackathon challenge? To enhance the personalization of their mobile app experience for users, to more effectively optimize their purchase conversion rates and profitability.
As is true of most startups, resources at Nelo are limited and roadmaps are endless. When Nelo hired their Head of Data Science, Alice Lu, her first task was to further improve their ML powered underwriting capabilities. Their data science team continues to focus on underwriting, fraud, pricing, and risk management. Meanwhile, after underwriting over 3 million transactions, Nelo had amassed a robust dataset on user demographics, product preferences, merchant repayment rates, and loan economics. It became clear to the Nelo team and to us that there was an exciting opportunity to harness this data to personalize the mobile app experience for each user in a way that could improve purchase conversion rates and profitability. It was also a problem perfectly suited for the expertise of Two Sigma technologists.
Over 3 days and 10 hours, we brought together a team of 20+ Two Sigma data scientists and engineers to tackle this data challenge head on. They were given a large dataset provided by Nelo and divided into small teams to create an algorithm that would ingest the code and output merchant suggestions for the user. At the end of the event, each team presented their thought process and technical approach to the Nelo data science team and engaged in Q&A. Those 200 person-hours condensed into 3 days the equivalent of about a month of a data scientist’s time, with the added benefit of the diversity of thought brought by people newly thinking about the problem, and the natural competitive and cooperative spirit that comes from working in teams.
In Alice’s words, “We had some high-level ideas about how we would do personalized recommendations, but it was inspirational to see how the engineers at Two Sigma came up with a diversity of creative yet practical solutions. We enjoyed our discussions with the teams, and truly felt fortunate to have this open platform that enables us to build something powerful together.” Kyle Miller, Co-Founder and CEO of Nelo, shared that “the results of this project will drive the way we leverage our data assets to personalize our app and optimize for profitability.”
Looking ahead, we know we can expect our portfolio companies to continue to generate large volumes of data that can create real value for their business. We look forward to working hand in hand with our founders to identify these opportunities and mobilize top technical talent to help them tackle them faster, cheaper, and “smarter together.”