GDAI: Cost-Effective AI-Driven Player Retention at ScaleGDAI is a cutting-edge platform that leverages AI to revolutionize player retention strategies for online casinos. In an industry where personalized engagement and timely interventions are critical, GDAI empowers CRM managers to optimize each player interaction—at scale. By integrating seamlessly with existing CRM systems, GDAI analyzes vast amounts of data to understand player behavior and preferences. This allows the platform to deliver real-time insights and personalized recommendations, ensuring every interaction enhances engagement, improves retention, and maximizes player lifetime value (LTV). With GDAI, online casinos can achieve cost-effective and scalable player retention, transforming data into actionable insights and driving sustainable growth.

As an AI consultant, I specialize in helping tech companies develop innovative products and lead teams through the design and implementation of intelligent systems. My approach blends technical know-how with practical business experience, ensuring tailored solutions for each client.Drawing from my roles as an Applied Scientist at Amazon and Head of Algorithms at Donde (now part of Shopify), I bring a comprehensive understanding of both the intricate details of technology and the overarching strategy required for leadership.I am passionate about problem-solving, combining intuition and logic to deliver the best possible solutions, always starting simple. My work spans the entire lifecycle of AI projects, from proof of concept to algorithm development, software implementation, analysis, and testing.The journey begins by translating business needs into AI problem definitions, followed by thorough literature review and designing the optimal algorithm. I then deploy production-ready code, train at the largest data scales, and serve millions of users.With a rich theoretical and practical knowledge of Machine Learning techniques, I tackle algorithms from end to end, applying them to real-world scenarios. My background includes individual contribution and management, focusing on the latest and most advanced models.My projects include developing a state-of-the-art face verification algorithm at Amazon, training a latent space representation for fashion to enhance customer navigation, deploying real-time image segmentation for a cutting-edge startup, and consulting on diverse topics such as video pose estimation, action recognition, online user behavior prediction, recommendation systems, and time-series sales forecasting.Proud to have collaborated with clients like Microsoft, GlobalDots, Shopper.ai, StartPlan, Rep, and many more, I bring a wealth of experience and a practical approach to every project.

At Amazon Lab126, an inventive research and development team responsible for high-profile consumer electronic devices, I had the privilege of contributing to some of Amazon's most advanced consumer products. My work included improving the Echo Look and Amazon Halo and engaging in groundbreaking projects within the “Smart Home” and “Smart Kitchen” realms.My role involved a blend of deep learning research and production-grade algorithmic coding, often leading projects from inception to deployment. I collaborated with multidisciplinary development teams, including Software, Data, System, and Hardware, to ensure the seamless integration and success of our initiatives.A highlight of my tenure was developing a state-of-the-art face verification algorithm, showcasing my ability to merge theoretical research with practical application. Additionally, I conducted extensive research in Computer Vision.

At Donde Search (now part of Shopify), I served as Head of Algorithms, focusing on research and development in Computer Vision and NLP using Deep Learning techniques. My role was pivotal in advancing the company's core engine technology.I built and managed the pipeline and Gantt chart for the core engine development, ensuring streamlined processes and timely delivery. My work included implementing multi-class/label classification on images using Transfer Learning and researching fashion landmark detection to enhance our product offerings.One of my significant contributions was the development and improvement of a multi-label text classification system, designed to be accessible for developers with varying levels of expertise. I also created a unique word embedding tailored to specific missions and built a pipeline for semi-automatic image annotations, significantly boosting classifier performance.My development efforts were primarily carried out using Keras and TensorFlow in Python, leveraging these powerful tools to push the boundaries of what our technology could achieve.

Researched Deep Learning algorithms for Vision problems in the e-commerce domain.Developed from pen-and-paper Real-Time Algorithms in the fields of Image Processing, Computer Vision, and Machine Learning including : Image Classification, Segmentation, Color models, Line Detection, Clustering, Tracking Objects, Bilateral Symmetry recognition, Instagram-like filters, Lighting issues, and many more.

- Real time Video Processing for an Underwater Robot- Responsible for writing the image processing algorithms- Simulations in Matlab- C++ Implementation using OpenCV libraries- Eclipse Environment- Linux OS