2016

Hi, I’m Robin, co-founder and CTO of Artie. I was 22 when I built my first ingestion pipeline at scale. I was working in a startup in San Francisco where we handled enormous amounts of data (billions of events per day). As a startup, we had to keep the pipelines extremely efficient to save time and money.

2018

The startup was acquired by Zendesk and I went on to build the foundations of Zendesk’s CRM offering. Change data capture (CDC) and Kafka was used heavily internally – it was such a big part of Zendesk that they developed and open sourced Maxwell, a CDC application that reads MySQL binlogs and writes data changes as JSON to Kafka, Kinesis, and other streaming platforms.

2021

I eventually went to work at a real estate tech startup in San Francisco, where I felt firsthand how data ingestion lag and high number of data processing errors led to lower productivity of teams downstream of the warehouse. I saw it directly lead to higher operational costs and impact on our ability to build robust data products. It's hard to build a house when you have to constantly worry about its foundation.

2022

Having used CDC in the past and knowing the benefits, I wanted to use it for data replication in order to minimize data latency. I tried a bunch of tools available in the market and was disappointed. So, I ended up architecting my own solution.

2022

I had an MVP working and it felt like magic! But, the process was hard and there was still so much more to build. It quickly dawned on me that every company would need to address these concerns in order to build a robust streaming CDC pipeline.  How do I guarantee eventual consistency? How do I guarantee all data types are supported? How do I handle schema evolution? How do I handle stateful datasets? How should backfills be handled to ensure minimal load and no storage growth? What about telemetry?

2023

I recruited my co-founder Jacqueline to help deal with everything else related to company building and go-to-market so I can focus on building the product. She got us into Y Combinator, got our first few dozen customers, and convinced world-class investors to join us on our journey. 

2024

We are even more excited in our mission to offer the fastest and most reliable database replication solution. As Joe Reis says, “The frequency of ingestion sets a ceiling on downstream frequency”. With Artie, our customers are now able to adopt all kinds of workloads that were previously unattainable, directly from their data warehouse.