Credit: Dreamstime
Cloud-based data warehouse company Snowflake has agreed to acquire artificial intelligence-based time series forecasting platform provider Myst AI.
“Time series forecasting is one of the most applied data science techniques in business. Accurate forecasts can establish measurements to guide management, facilitate planning and goal setting, and help mitigate risk,” Greg Czajkowski, senior vice president of engineering and support at Snowflake, wrote in a blog post announcing the deal.
The technique is used in supply chain management, inventory planning and finance, and public health, Czajkowski said, adding that in manufacturing it can be used to predict product demand at a geographic level to help meet shipping requirements and reduce inventory waste.
Time series forecasting looks for statistical patterns in past data to generate future predictions as an aid to strategic decision making.
Myst AI’s focus is on applying it to reduce energy costs by forecasting spikes in demand, renewable generating capacity, and prices. The acquisition, according to Czajkowski, is a continuation of Snowflake’s strategy to focus on building machine learning extensibility into its data cloud.
In June 2022 the company announced several new features, including Streamlit integration and machine learning for SQL users, in an effort to further its machine learning strategy.
Snowflake acquired Python-based Streamlit last year to help machine learning and data science engineering teams to visualise, mutate, and share data.
Myst AI has raised $8 million in venture funding since 2020, but Snowflake did not say how much it paid for the company. This now marks Snowflake’s sixth acquisition in three years.
In August 2022 it bought Polish AI-based document understanding platform Applica to help enterprises handle unstructured data.
The others were Streamlit (March 2022), Polish custom software company Pragmatists (January 2022), Polish digital products development studio Polidea (February 2021) and Canadian data anonymisation company CryptoNumerics (July 2020).
