TL;DR
Get monitors, keyboards and dev gear delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
The Pragmatic Engineer published an interview with Peter Mattis, Cockroach Labs’ co-founder and CTO, about the engineering behind distributed databases and storage systems. Mattis also discusses how AI tools have brought him back to writing code and why he believes experienced engineers can use them to improve their output.
The Pragmatic Engineer has published an interview with Peter Mattis, co-founder and CTO of Cockroach Labs, about how engineers build storage and database systems that remain fast, reliable and correct at scale. Mattis, who also co-created the image editor GIMP and worked on Gmail and distributed storage at Google, discusses the design choices behind those systems and says AI tools have brought him back to writing code after his work shifted toward management.
The conversation traces Mattis’s work from open-source software through Google to Cockroach Labs. Among the examples in the accompanying report are Gmail’s early storage design, where B-trees tracked email threads and unread counts, and Colossus, Google’s successor to the Google File System. The report says Colossus used Reed–Solomon erasure coding to store data twice while increasing redundancy, compared with the three full copies used by the earlier system.
Mattis also describes replacing standard data structures when their performance or memory use became a concern. At Google, he built a B-tree with semantics similar to C++’s std::map, which the report says used less memory and ran faster because of spatial locality. He later developed a Swiss Table implementation for Go that was faster than the language’s existing map implementation; the report says it was subsequently incorporated into Go’s standard library with help from the Go team.
The interview turns to AI-assisted programming as well. The source describes Mattis as having regularly shipped roughly 100,000 lines of database-grade code per year before AI and says he now feels more productive without a drop in quality. That is his reported experience, not a published measurement or independent assessment. He argues that AI can improve quality and increase the impact of engineers with deep knowledge of their systems. The discussion also considers how code review may change as agents produce more work.
How Storage Choices Shape Reliability
The examples show why database engineering involves tradeoffs across speed, storage overhead and redundancy. Colossus’s reported use of erasure coding is a concrete illustration: changing how data is represented can reduce the amount stored while maintaining or improving redundancy. For systems serving large amounts of data, such design decisions affect both operating costs and resilience.
The interview also connects low-level implementation choices to performance at scale. B-trees appear in the report across Gmail’s storage, Google data structures and CockroachDB’s range index. Mattis’s account offers engineers and readers a practitioner’s view of how established structures can be adapted to particular workloads. His comments on AI raise a separate question for software teams: how to review and maintain code when experienced developers use agents to produce more of it. The interview shares his view, but does not establish how widely that experience applies.
Top picks for "distribut databas peter"
As an affiliate, we earn on qualifying purchases.
From GIMP to Google Storage
Mattis and Spencer Kimball created GIMP while they were college students. The report says the first version of the Google logo was made with the image editor. In 2001, Sergey Brin invited Mattis to interview at Google and made him an offer, which Mattis initially declined because of the commute from San Francisco to Mountain View. He joined another startup before later accepting a second opportunity at Google.
At Google, Mattis worked on Gmail and distributed storage. Gmail launched in 2004 with 1GB of free email storage, a figure that seemed unusually generous to many users at the time. The report links B-trees to the service’s handling of email threads. It also describes Colossus as a later step in Google’s storage infrastructure. Mattis subsequently co-founded Cockroach Labs, the company behind the distributed SQL database CockroachDB.
The source is an interview and accompanying episode report, rather than a new technical paper or company announcement. Its descriptions of past engineering work and Mattis’s views are attributed to the conversation; the material provided does not include independent testing of the performance claims.
““There’s always going to be someone else working on your idea. You can’t get dissuaded if they pre-announce it.””
— Peter Mattis, as quoted in The Pragmatic Engineer report
What the Interview Does Not Measure
The source material does not give a publication date, detailed methodology or independent evaluation for Mattis’s account of coding productivity. It does not define how the reported annual code volume was counted, how quality was assessed, or which AI tools and tasks were involved. The claim that he feels more productive without a quality decline should therefore be read as his personal experience, not as a general finding about AI-assisted development.
The provided report also offers only selected technical details about Gmail and Colossus, rather than full design documentation or benchmarks. It does not quantify the systems’ performance gains beyond the stated storage and redundancy comparison. The interview raises questions about the future of code review but does not set out a specific process or predict when teams will adopt one.
Further Listening and Technical Detail
The Pragmatic Engineer directs readers to the full episode and transcript, with listening options on YouTube, Apple and Spotify and timestamps listed with the episode. Those materials may provide more detail on Mattis’s explanations of database design, engineering work at Google and AI-assisted coding than the summary report alone.
For readers following the storage discussion, the report points to further material on low-latency routing in space and a paper titled “The Ubiquitous B-Tree.” It does not announce a product release, new Cockroach Labs feature or scheduled technical milestone. Any broader claims about AI’s effects on engineering practice remain open for further evidence and reporting.
Key Questions
What is the news development?
The Pragmatic Engineer published an interview with Peter Mattis about distributed databases, storage systems and AI-assisted software development.
Who is Peter Mattis?
Mattis is co-founder and CTO of Cockroach Labs. He also co-created GIMP and worked on Gmail and distributed storage at Google, according to the report.
What does the report say about Colossus?
The report says Google’s Colossus storage system used Reed–Solomon erasure coding to store data twice while increasing redundancy, compared with three full copies in the earlier Google File System.
What does Mattis say about AI coding tools?
Mattis says AI has made him feel more productive and brought him back to writing code. The source presents this as his account; it does not provide independent measurements of productivity or code quality.
Source: rss
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
