Professor Andy Pavlo, a world reference in database systems, announced that he is joining ClickHouse to create and lead ClickHouse Labs, a new research group at the company. The news was shared on ClickHouse’s official blog.
Pavlo has been a professor in the Computer Science Department at Carnegie Mellon University since 2013. Throughout his career, he has focused on understanding the internals of modern database management systems (DBMS), closely following every new system that emerges, both in industry and academia.
The researcher has known ClickHouse since its release as open-source software in June 2016. At the time, his initial reaction was that it was ‘vaporware,’ as it seemed too good to be true. ClickHouse already had features that only existed in a few commercial and closed analytical DBMSs.
One of the differentiators was its implementation in C++ with vectorized execution using SIMD, something rare in open systems that year. Most open-source analytical DBMSs in 2016 were JVM-based and only adopted SIMD optimizations years later.
Since then, Pavlo has closely followed ClickHouse’s evolution, which has always stood out as a leading system and relevant to his academic research projects. He recalls wearing an original ClickHouse t-shirt in his first remote classes in 2020, when the pandemic forced database courses to move online.
The invitation to found ClickHouse Labs came from the company’s co-founders, and Pavlo did not hesitate to accept. He compares the collaboration with ClickHouse’s engineering team to a heavyweight partnership in hip-hop, like the union of Killer Mike and El-P to form a supergroup.
The goal of ClickHouse Labs is to create a high-level industrial research organization focused on databases. The idea is not to be an isolated group that merely ‘throws ideas over the wall’ to engineering. On the contrary, the lab will work closely with engineers, customers, contributors, and industry partners to develop and disseminate new ideas that keep ClickHouse at the forefront.
The lab will also work with ClickHouse’s PostgreSQL team to consolidate the company’s managed service as a benchmark in performance and reliability. The combination of PostgreSQL and ClickHouse, which serve different workloads, offers a broad foundation to investigate transactional and analytical problems.
The mission is clear and ambitious: to conduct research with scientific value and turn the best ideas into technology that makes a difference for users. Pavlo wants to achieve the same level of impact as pioneering organizations like IBM Research and Microsoft Research, which showed that industrial labs can advance computer science, influence commercial products, and train generations of researchers.
ClickHouse already has a notable history of detailed technical publications. Since the company’s founding in 2021, engineers have produced in-depth articles on the DBMS implementation. There is also the VLDB 2024 paper describing ClickHouse’s core architecture. These works are so complete that Pavlo uses them as required reading for his students at Carnegie Mellon.
One of Pavlo’s immediate priorities is to accelerate the validation of ideas and optimizations that the engineering team has already explored but hasn’t had time to bring to production. This will be the starting point to explore new frontiers and further drive ClickHouse.
A central question the lab will investigate is how DBMSs like ClickHouse and PostgreSQL fit into emerging AI and agent technologies. The problem has two sides: first, defining what a DBMS should be to best support agents; second, how agents can improve and automate the development of DBMSs themselves.
Everything is open: new hardware, algorithms, data structures, execution strategies, and new ways to build and operate database software. Pavlo admits he doesn’t yet have answers to these questions — after all, that’s why it’s research — but he is confident that ClickHouse’s solid relational model foundation positions it well to evolve with these data-intensive workloads.
With ClickHouse Labs, Pavlo sees the opportunity to create an organization dedicated to advancing knowledge in databases, training new researchers, and developing technology that benefits the entire community.