“ DigitalTurbine tackled the challenge of optimizing ad bids in real-time auctions. Using advanced techniques like contextual bandits with Thompson Sampling, we crafted a dynamic system that learns from market trends and other bidders' decisions. This empowered DigitalTurbine to make lightning-fast, smart decisions, ensuring optimal ad placements and boosting competitiveness in the advertising landscape. ”

“ Unity Aura is a news feed that provides users with personalized content recommendations. Argmax took part in desigining and developing the recommendation engine, which was based on a hybrid approach of vector similarity and session based recommendations. The system supports multi-lingual articles from a variety of sources, and is currently serving thousands of users on a daily basis. ”

“ MoonActive transformed CoinMaster's in-game economy by implementing a two-step causal approach. First, a life-time-value (LTV) model predicted player value over time. Then, leveraging the LTV model, we optimized coin sales and promotions, maximizing player engagement, minimizing churn, and achieving a balanced and profitable in-game economy. ”

“ Visa empowered local businesses with tailored recommendations using an innovative model that analyzed transaction history to suggest similar businesses aligned with customer interests. With privacy as a priority, we processed data through a trained word2vec model and a vector similarity server. This led to a remarkable 150% increase in sales, showcasing Visa's commitment to fostering thriving communities and strengthening local economies.”