Our Investment in MindsDB
Democratizing Machine Learning.
What do Bacon and Machine Learning have in common?
Francis Bacon, the 15th century English philosopher and statesman, has been called the father
of empiricism for his contributions to the development of the scientific method. In philosophy he
is also widely credited with the story about adjusting position when “if the mountain won’t come
to you, you go to the mountain”. At Walden Catalyst Ventures we believe in both the scientific
method for machine learning and there is greater advantage to bringing machine learning
models to the ‘mountains’ of data than the data to the models. Today's enterprises are
increasingly leveraging existing corporate data to transform their business decision making
processes. They are looking to use machine learning based predictions and need less
expensive and more accessible access to machine learning models without excessive data
preparation. Seeing the incredible opportunity to deploy machine learning at scale within the
enterprise is why we’re partnering with and investing in MindsDB as they build a category
defining company and democratize machine learning.
Just as Moore’s law in the semiconductor industry led to an explosion in innovation of compute
intensive applications, the availability of data and innovation in machine learning models is
leading to an explosion in enterprises leveraging machine learning predictions to drive better
business decisions. The global machine learning market size was $US11.3B in 2020 with
machine learning solutions and service companies experiencing positive demand shocks during
the COVID pandemic. Based on analysis from Fortune Business Insight the global machine
learning market also experienced significant growth of 36.2% during 2020 and is projected to
grow from US$15.5B in 2021 to $USD152.3B by 2028 at a compound annual growth rate
(CAGR) of 38.6%, attributable to advancements in machine learning technology and integration
of machine learning capabilities within data-driven solutions. Many e-commerce companies
such as Amazon, Alibaba, and Shopify are leveraging machine learning to boost their sales and
enhance customer experiences while financial institutions from banking, insurance, Wall Street
to Main Street, are leveraging machine learning for algorithmic training, credit scoring, anomaly
detection, and digital wealth management. So, how are enterprises currently attempting to
leverage existing corporate data to transform their businesses using machine learning based
Unfortunately, specialized machine learning expertise continues to be expensive and difficult for enterprises to hire, justify, and retain. Early stage companies that have attempted to address these challenges over the last few years have found their current solutions require substantial amounts of data preparation, cleaning, and labelling, plus hard to find machine learning/AI data scientists to conduct feature engineering; build, train, and optimize models; assemble, verify, and deploy into production; and then monitor in real time, improve, and refine. A recent study has shown it takes 57% of companies a month, to over a year, to deploy a machine learning model into production. Leveraging existing databases and automating the feature engineering, building, training, and optimization of models, assembling them, and deploying them into production is called AutoML and has been gaining traction within enterprises for enabling non-experts to use machine learning
models for practical applications. If the operational cost to design and implement machine learning models is prohibitive to the project then we must examine viable alternatives of bringing the models closer to the data.
With the recent rapid advances in analytics and machine learning it’s taken time for business databases and systems of record to ‘catch-up’, connect, and leverage the potential power these combined platforms present. Machine learning models require multiple iterations with existing data to train. Additionally, extracting, transforming, and loading (ETL) data from one system to another is complicated, leads to multiple copies of information, and is a compliance and tracking nightmare. What’s smarter is to bring machine learning models to the data, interacting with these models with standard SQL commands, automating those models so they are constantly up-to-date, and avoiding all aforementioned issues. The other advantage of bringing machine learning models to the database is leveraging the existing skills, knowledge, and experience of your present SQL database developers who will quickly come up the learning curve to reach productivity faster. How do contemporary enterprise customers leverage the power of their existing databases using MindsDB?
MindsDB brings machine learning to existing SQL databases with a concept called AI-Tables. AI-Tables integrate the machine learning models as virtual tables inside a database, creates a prediction, and can be queried with simple SQL statements. Almost instantly, time series, regression, and classification predictions can be done directly in your database and credit scoring, anomaly detection, modelling, segmentation, product recommendation, predictive maintenance, optimization of supply/demand/price/margin, and time-series forecasting, which presents a challenge even to seasoned ML engineers because of the often high degree of cardinality become as simple as executing a SQL query. MindsDB has partnered with the world's most scalable, multi-cloud databases, such as Clickhouse, DataStax (built on Apache Cassandra™), Confluent (built on Apache Kafka™), MariaDB, MongoDB, MySQL, PostgreSQL, SingleStore, and Snowflake to connect MindsDB’s machine learning platform to their respective data stores and enables your database to become a predictive engine while greatly reducing the time, cost, and complexity of operationalizing machine learning.
Together, Walden Catalyst Ventures and MindsDB believe in both the scientific method for machine learning and the advantage to bringing machine learning models to the ‘mountains’ of data. We’ve reviewed how today's enterprises are looking to use machine learning based predictions, need less expensive and more accessible access to machine learning models without excessive data preparation, and we believe MindsDB is the answer to those challenges.
To see how MindsDB can help you visit www.mindsdb.com.