December 2020 Summaries

Tecton and Amazon SageMaker feature stores are designed to address the challenges of building and serving high-quality machine learning features in training and production environments. A complete feature store should maximize data utility, enable self-sufficient data scientists, provide easy access to accurate historical data, improve collaboration on features, ensure high performance, especially in serving, and enable effective governance. Tecton's feature store offers a wide range of capabilities, including batch, streaming, and real-time data support, automated transformations, online and offline storage, training datasets with time travel, sharing and discovery of features, enterprise SLAs, monitoring, and features as code with full feature lineage. In contrast, SageMaker's feature store lacks some key capabilities, such as real-time or streaming data transformations, intelligent handling of feature versions, and built-in monitoring capabilities. Ultimately, the choice between Tecton and SageMaker depends on the specific needs of your organization and the scope of their feature management requirements.

Today marks a significant milestone for Tecton as its feature store is now in General Availability and has secured $35 Million in Series B funding to fuel its next phase of growth. The company was founded with the vision to make it easy and safe to build smart product experiences with machine learning, enabling every team to build with the speed, trust, and power of industry-leading applied ML organizations. Tecton's feature store aims to solve the problems of data scientists spending time cleaning and engineering features, and instead provides an enterprise-grade solution that combines minimal SDKs, end-to-end management, built-in best practices, governance workflows, security, scalability, and delivery as a cloud-native service. With this funding, Tecton plans to build on its roadmap, bringing the feature store to all major clouds, supporting broader data processing frameworks, and simplifying the feature engineering experience for data scientists.