November 2022 Summaries

The Tecton apply() conference discussed various themes related to machine learning and data engineering, including productionization, data architectures for machine learning applications, real-time machine learning, rapid model deployment in healthcare, best practices for scaling a company's data team, and improving collaboration between engineering and data teams. The event highlighted the challenges of putting good models into production, the importance of unified data models, and the need for faster iteration cycles and better ways to manage data. It also showcased successful real-time machine learning applications, such as CashApp's personalization experiences and ERAdvisor's emergency room wait time estimates. Additionally, speakers shared best practices for scaling data teams and improving collaboration between engineering and data teams as companies mature. The next event, apply(recsys), will take place on December 6 and will focus on recommender systems.

Our blog series, Featured Tectonauts, highlights the amazing characters we get to interact with daily here at Tecton. I work as a Product Manager at Tecton supporting our Compute, Real-Time Serving, Data Observability and DevOps engineering teams, helping them build impactful products for customers. I was pleasantly surprised by the technical strength of our engineering team, which is higher than expected in a startup. Outside of work, I relieve stress by throwing balls with my two-year-old Golden Retriever, Maple. Interestingly, we can fit 9 helmets and various sports gear in our garage. The book "Why We Sleep" influences me to sleep more every day. I've taken long train rides, including a 14-hour flight to Tokyo and 2 hours to Sapporo for snowboarding.

HelloFresh has selected Tecton as its feature store to standardize machine learning across the organization, aiming to create a data layer that can be trusted by all teams. The company benchmarked various solutions and found Tecton fulfilled 90% of the feature store requirements compared to other vendors. Tecton will help HelloFresh streamline features from different data sources, ensure data quality and consistency, and supervise training pipelines to avoid common ML pitfalls. This will enable the company to build robust predictive products and automate decision-making across its value chain.