Todd Beauchene is an experienced technology leader and architect specializing in data warehousing and business intelligence. He has held significant roles at companies like Snowflake and Alation, and is currently working at Revefi. Todd holds a degree from Brigham Young University.
This past December, I joined Revefi and I wanted to share what led to that decision. I have had the good fortune to work for a number of great companies within the data industry and in that time I have learned a great deal. For a company to break through the crowded market and have a real and lasting impact, it needs to address a real problem that organizations face and are willing to pay you to solve. On top of that, you need a top-notch team with a singular focus to create a differentiated solution.
Over the past decade, the data industry has been transformed by public cloud computing. The best-run organizations have always leveraged data to make decisions that allow them to outmaneuver their competition. The systems required to collect and analyze this data have traditionally been extremely expensive and were limited in scale. The cloud data platforms developed since 2010 have marked a significant shift in how data is stored and processed.
As with most cloud computing, one of the compelling features of cloud data platforms is consumption-based pricing. This feature allows organizations that could never afford a Data Warehouse appliance to leverage the power of a Cloud Data Warehouse for a few dollars an hour. But this new cost structure comes with a whole new set of challenges.
Revefi aims to solve the question of how to optimize a Cloud Data Platform. It approaches this question not only from a cost perspective but also by leveraging its founding team’s unique experience with query and platform optimization to deliver a holistic solution for the leading platforms.
The Rise of the Cloud Data Platform
A little over 10 years ago I was offered the opportunity to join Snowflake, a company that very few people even knew existed at the time. I was working as a BI Engineer at Amazon at the time and loved working with the complex datasets related to the world’s largest online marketplace, but this tiny startup had technology that really intrigued me. I was knee-deep in trying to scale a few data projects within the mobile shopping business unit, and my first thought when I saw a demo of Snowflake was that I wished I could use it in my current job.
What really stood out to me about Snowflake’s technology was the way it separated storage from compute in order to allow multiple, independent compute clusters to operate on a single copy of the data without any resource contention. I was used to working with databases where DBAs constantly had to monitor workloads and resource contention could bring the data warehouse to a crawl at times.
One of Snowflake’s most significant innovations was leveraging S3 storage in a unique way. I feel that S3 was perhaps the most important technology developed by AWS and many startups were exploring ways to leverage its storage for a whole range of applications. Snowflake built an interface that felt like a standard MPP SQL Database, but stored data in a way different from any database that came before. When I joined the company, there were still a lot of things to figure out, but the storage model was set from very early on.
Now Snowflake wasn’t the only company working on a cloud data platform. Google launched BigQuery in 2011 and AWS launched Redshift a year later. Databricks launched in 2015, within a month of the Snowflake launch. While these represented 4 very different technology visions, the 4 have converged in unexpected ways as each has developed into a cloud data platform.
The Decision
Last summer I attended the Snowflake Summit, and as you can imagine I ran into a lot of old friends and acquaintances. One of those was Sanjay Agrawal, who was manning a booth for his new startup named Revefi. I had worked with Sanjay at ThoughtSpot and respected his contributions as a co-founder and enjoyed collaborating with him. As we spoke, I was surprised at the depth of his knowledge of the inner workings of Snowflake and intrigued by what they were building.
Leveraging the usage logs and other platform metadata, Revefi is able to provide observability and data quality on top of cost optimization. And on top of that, Revefi had recently rolled out RADEN, the Revefi AI Data Engineer which leverages generative AI to help customers understand their cloud data platform even better.
A few months ago, it became clear that it was time to find my next opportunity. I remembered that conversation and so I reached out to see if there was a role available that would fit me. This led to deeper discussions where Sanjay shared what they had added to the product and what they were working on. In subsequent days, I had the opportunity to meet the other members of the team and I came to the conclusion that Revefi had assembled a team capable of building and delivering a solution with real value. So I took the plunge and decided to join as Head of Growth within the Engineering organization, a unique role that I feel allows me to leverage my expertise in data platforms and my affinity for working directly with customers.
I feel strongly that any organization that is currently leveraging a Cloud Data Platform would benefit from Revefi. Whether you want to control costs, increase trust in data, or optimize queries for your specific platform; one tool can do it all.
The Post-Modern Data Stack
I had the good fortune to spend a number of years within the Alliances organization at Snowflake at a critical time when the cloud data ecosystem began to take shape. I worked closely with the technologies that would come to be referred to as the Modern Data Stack (we actually called it the Dream Data Stack in those days). These technologies collected data from a wide variety of sources and brought it into the cloud data platforms. These scalable platforms offered an efficient way to then transform the raw data into a business-friendly schema that allowed for easy analysis, with part of the stack orchestrating this data transformation. Finally, Business Intelligence tools made this data available to users at every level of the business in order to help them make decisions.
While the Modern Data Stack offers a simple way to leverage data within a business of any size, I would argue that we have arrived at a point in time where a new stack is needed. As these platforms continue to evolve, new ways of processing and accessing data require a different set of tools. Artificial Intelligence garners many of the headlines, but observability and data quality are also essential parts of this new stack.
These new tools need to understand that organizations have multiple environments and even multiple data platforms running within their departments. They must understand the inner workings of each platform, and even be able to compare and contrast the strengths of each. On top of that, this new stack should leverage the latest ML and AI techniques and automate as many tasks as possible. I am not sure of all the layers of this new stack, but I have placed my bet that Revfi is going to be a part of it.
The best way to see what Revefi can do for your organization is to see it in action, running against your own platform(s). You can try it for yourself by registering on our website. Please reach out to me if you want more details.
