Bringing Rust into the Linux kernel is one of the most ambitious modernization efforts in open source history. The Linux kernel, with its decades of C code and deeply ingrained development practices, is now opening its doors to a memory-safe language. It's the first time in over 30 years that a new programming language has been officially adopted for kernel development. But the journey is far from straightforward.
In this episode, we speak with Danilo Krummrich, Linux kernel maintainer and Rust for Linux core team member, about the groundbreaking work of integrating Rust into the Linux kernel. Among other things, we talk about the Nova GPU driver, a Rust-based successor to Nouveau for NVIDIA graphics cards, and discuss the technical challenges and cultural shifts required for large-scale Rust adoption in the kernel as well as the future of the Rust4Linux project.
About Rust for Linux
Rust for Linux is a project aimed at bringing the Rust programming language into the Linux kernel. Started to improve memory safety and reduce vulnerabilities in kernel code, the project has been gradually building the infrastructure, abstractions, and tooling necessary for Rust to coexist with the kernel's existing C codebase.
About Danilo Krummrich
Danilo Krummrich is a software engineer at Red Hat and a core contributor to the Rust for Linux project. His fundamental contribution to Rust for Linux is the driver-core infrastructure, the foundational framework that makes it possible to write drivers in Rust at all. This includes both C and Rust code that provides the core abstractions for device drivers in the kernel. Danilo is a maintainer for multiple critical kernel subsystems, including Driver Core, DRM (GPUVM, Rust, GPU Scheduler), GPU drivers for NVIDIA GPUs (Nova, Nouveau), Firmware Loader API, as well as Rust bindings for PCI, DMA, and ALLOC. He is the primary developer of the Nova GPU driver, a fully Rust-based driver for modern NVIDIA GPUs.
Links From The Episode

• AOSP - The Android Open Source Project
• Kernel Mailing Lists - Where the Linux development happens
• Miguel Ojeda - Rust4Linux maintainer
• Wedson Almeida Filho - Retired Rust4Linux maintainer
• noveau driver - The old driver for NVIDIA GPUs
• Vulkan - A low level graphics API
• Mesa - Vulkan and OpenGL implementation for Linux
• vtable - Indirect function call, a source of headaches in nouveau
• DRM - Direct Rendering Manager, Linux subsystem for all things graphics
• Monolithic Kernel - Linux' kernel architecture
• The Typestate Pattern in Rust - A very nice way to model state machines in Rust
• pinned-init - The userspace crate for pin-init
• rustfmt - Free up space in your brain by not thinking about formatting
• kunit - Unit testing framework for the kernel
• Rust core crate - The only part of the Rust Standard Library used in the Linux kernel
• Alexandre Courbot - NVIDIA employed co-maintainer of nova-core
• Greg Kroah-Hartman - Linux Foundation fellow and major Linux contributor
• Dave Airlie - Maintainer of the DRM tree
• vim - not even neovim
• mutt - classic terminal e-mail client
• aerc - a pretty good terminal e-mail client
• Rust4Linux Zulip - The best entry point for the Rust4Linux community

Official Links

• Rust for Linux GitHub
• Danilo Krummrich on GitHub
• Danilo Krummrich on LinkedIn

What does it take to rewrite the foundational components of one of the world's most popular Linux distributions? Ubuntu serves over 12 million daily desktop users alone, and the systems that power it, from sudo to core utilities, have been running for decades with what Jon Seager, VP of Engineering for Ubuntu at Canonical, calls "shaky underpinnings."
In this episode, we talk to Jon about the bold decision to "oxidize" Ubuntu's foundation. We explore why they're rewriting critical components like sudo in Rust, how they're managing the immense risk of changing software that millions depend on daily, and what it means to modernize a 20-year-old operating system without breaking the internet.
About Canonical
Canonical is the company behind Ubuntu, one of the most widely-used Linux distributions in the world. From personal desktops to cloud infrastructure, Ubuntu powers millions of systems globally. Canonical's mission is to make open source software available to people everywhere, and they're now pioneering the adoption of Rust in foundational system components to improve security and reliability for the next generation of computing.
About Jon Seager
Jon Seager is VP Engineering for Ubuntu at Canonical, where he oversees the Ubuntu Desktop, Server, and Foundations teams. Appointed to this role in January 2025, Jon is driving Ubuntu's modernization strategy with a focus on Communication, Automation, Process, and Modernisation. His vision includes adopting memory-safe languages like Rust for critical infrastructure components. Before this role, Jon spent three years as VP Engineering building Juju and Canonical's catalog of charms. He's passionate about making Ubuntu ready for the next 20 years of computing.
Links From The Episode

• Juju - Jon's previous focus, a cloud orchestration tool
• GNU coretuils - The widest used implementation of commands like ls, rm, cp, and more
• uutils coreutils - coreutils implementation in Rust
• sudo-rs - For your Rust based sandwiches needs
• LTS - Long Term Support, a release model popularized by Ubuntu
• coreutils-from-uutils - List of symbolic links used for coreutils on Ubuntu, some still point to the GNU implementation
• man: sudo -E - Example of a feature that sudo-rs does not support
• SIMD - Single instruction, multiple data
• rust-coreutils - The Ubuntu package with all it's supported CPU platforms listed
• fastcat - Matthias' blogpost about his faster version of cat
• systemd-run0 - Alternative approach to sudo from the systemd project
• AppArmor - The Linux Security Module used in Ubuntu
• PAM - The Pluggable Authentication Modules, which handles all system authentication in Linux
• SSSD - Enables LDAP user profiles on Linux machines
• ntpd-rs - Timesynchronization daemon written in Rust which may land in Ubuntu 26.04
• Trifecta Tech Foundation - Foundation supporting sudo-rs development
• Sequioa PGP - OpenPGP tools written in Rust
• Mir - Canonicals wayland compositor library, uses some Rust
• Anbox Cloud - Canonical's Android streaming platform, includes Rust components
• Simon Fels - Original creator of Anbox and Anbox Cloud team lead at Canonical
• LXD - Container and VM hypervisor
• dqlite - SQLite with a replication layer for distributed use cases, potentially being rewritten in Rust
• Rust for Linux - Project to add Rust support to the Linux kernel
• Nova GPU Driver - New Linux OSS driver for NVIDIA GPUs written in Rust
• Ubuntu Asahi - Community project for Ubuntu on Apple Silicon
• debian-devel: Hard Rust requirements from May onward - Parts of apt are being rewritten in Rust (announced a month after the recording of this episode)
• Go Standard Library - Providing things like network protocols, cryptographic algorithms, and even tools to handle image formats
• Python Standard Library - The origin of "batteries included"
• The Rust Standard Library - Basic types, collections, filesystem access, threads, processes, synchronisation, and not much more
• clap - Superstar library for CLI option parsing
• serde - Famous high-level serilization and deserialization interface crate

Official Links

• Canonical
• Ubuntu
• Jon Seager's Website
• Jon's Blog: Engineering Ubuntu For The Next 20 Years
• Canonical Blog
• Ubuntu Blog
• Canonical Careers: Engineering - Apply your Rust skills in the Linux ecosystem

Building a new programming language from scratch is a monumental undertaking. In this episode, we talk to Richard Feldman, creator of the Roc programming language, about building a functional language that is fast, friendly, and functional. We discuss why the Roc team moved away from using Rust as a host language and instead is in the process of migrating to Zig. What was the decision-making process like? What can Rust learn this decision? And how does Zig compare to Rust for this kind of systems programming work?
About Roc
Roc is a fast, friendly, functional programming language currently in alpha development. It's a single-paradigm functional language with 100% type inference that compiles to machine code or WebAssembly. Roc takes inspiration from Elm but extends those ideas beyond the frontend, introducing innovations like platforms vs applications, opportunistic mutation, and purity inference. The language features static dispatch, a small set of simple primitives that work well together, and excellent compiler error messages. Roc is already being used in production by companies like Vendr, and is supported by a nonprofit foundation with corporate and individual sponsors.
About Richard Feldman
Richard Feldman is the creator of the Roc programming language and author of "Elm in Action." He works at Zed Industries and has extensive experience with functional programming, particularly Elm. Richard is also the host of Software Unscripted, a weekly podcast featuring casual conversations about code with programming language creators and industry experts. He's a frequent conference speaker and teacher, with courses available on Frontend Masters. Richard has been a longtime contributor to the functional programming community and previously worked at NoRedInk building large-scale Elm applications.
Links From The Episode

• Zig - Better than Rust?
• Rust in Production: Zed - Our interview with Richard's colleague with more details about Zed
• Richards blogpost about migrating from Rust to Zig - Sent in by many listeners
• Elm - Initial inspiration for Roc
• NoRedInk - Richard's first experience with Elm
• Haskell - A workable Elm on the backend substitute
• OCaml - Functional language, but pure functions only encouraged
• F# - Similar shortcomings as OCaml
• Evan Czaplicki - Creator of Elm
• Ghostty - Terminal emulator from Mitchel Hashimoto with lots of code contributions in Zig
• bumpref - A tiny Rust crate that came out of this discussion, providing Arc::bump(), which is an alias for clone().
• RAII - Resource acquisition is initialization, developed for C++, now a core part of Rust
• Frontend Masters: The Rust Programming Language - Richard's course teaching Rust
• Rust by Example: From and Into - Traits for ergonomic initialising of objects in Rust
• The Rust Programming Language: Lifetime Annotations on Struct Definitions - Learn from Roc: try to avoid having lifetime type parameters
• Rust By Example: Box, stack and heap - Putting objects on the heap can slow down your application
• Design Patterns: Elements of Reusable Object-Oriented Software - Seminal book popularising many common patterns in use today, written by the so-called "Gang of Four"
• Casey Muratori: The Big OOPs - Game developer explaining why OOP was an obvious mistake for high performance code
• Alan Kay - Coined the term "object-oriented" while developing the Smalltalk language in the 70s
• Niklaus Wirth - Working on Modula, a modular programming language, at the same time
• Kotlin - A new and popular language, basically Java++
• Go - Popular "greenfield" language, i.e. not coupled to an existing language, not using the object oriented paradigm
• Cranelift backend for Rust - A faster backend than LLVM, but still not released
• Andrew Kelly - Creator of Zig
• Software Unscripted - Richard's Podcast
• GPUI - Zed's own UI crate
• Structure of Arrays vs Array of structures - A big source of unsafe code in the Rust implementation of Roc
• The Zig Programming Language: comptime - Zig's replacement for Rust's proc-macros, with much broader utility
• crabtime - Comptime crate for Rust
• Roc - Roc's namesake, the mythical bird
• Rust in Production: Tweede Golf - Podcast episode with Volkert de Vries, one of the first contributors to Roc

Official Links

• Roc Programming Language
• Roc on GitHub
• Richard Feldman on GitHub
• Richard Feldman on LinkedIn
• Richard Feldman on X
• Software Unscripted Podcast

How do you build a system that handles 90 million requests per second? That’s the scale that Cloudflare operates at, processing roughly 25% of all internet traffic through their global network of 330+ edge locations.

In this episode, we talk to Kevin Guthrie and Edward Wang from Cloudflare about Pingora, their open-source Rust-based proxy that replaced nginx across their entire infrastructure. We’ll find out why they chose Rust for mission-critical systems handling such massive scale, the technical challenges of replacing battle-tested infrastructure, and the lessons learned from “oxidizing” one of the internet’s largest networks.
About Cloudflare

Cloudflare is a global network designed to make everything you connect to the Internet secure, private, fast, and reliable. Their network spans 330+ cities worldwide and handles approximately 25% of all internet traffic. Cloudflare provides a range of services including DDoS protection, CDN, DNS, and serverless computing—all built on infrastructure that processes billions of requests every day.

About Kevin Guthrie

Kevin Guthrie is a Software Architect and Principal Distributed Systems Engineer at Cloudflare working on Pingora and the production services built upon it. He specializes in performance optimization at scale. Kevin has deep expertise in building high-performance systems and has contributed to open-source projects that power critical internet infrastructure.

About Edward Wang

Edward Wang is a Systems Engineer at Cloudflare who has been instrumental in developing Pingora, Cloudflare’s Rust-based HTTP proxy framework. He co-authored the announcement of Pingora’s open source release. Edward’s work focuses on performance optimization, security, and building developer-friendly APIs for network programming.

Links From The Episode

• Pingora - Serving 90+ million requests per second (7e12 per day) at Cloudflare
• How we built Pingora - Cloudflare blog post on Pingora’s architecture
• Open sourcing Pingora - Announcement of Pingora’s open source release
• Rust in Production: Oxide - Interview with Steve Klabnik
• Anycast - Routing traffic to the closest point of presence
• Lua - A small, embeddable scripting language
• nginx - The HTTP server and reverse proxy that Pingora replaced
• coredump - File capturing the memory of a running process for debugging
• OpenResty - Extending nginx with Lua
• Oxy - Another proxy developed at Cloudflare in Rust
• Ashley Williams - Famous Rust developer who worked at Cloudflare at one point
• Yuchen Wu - One of the first drivers of Pingora development
• Andrew Hauck - Early driver of Pingora development
• Pingora Peak - The actual mountain in Wyoming where a Cloudflare product manager almost fell off
• shellflip - Graceful process restarter in Rust, used by Pingora
• tableflip - Go library that inspired shellflip
• bytes - Reference-counted byte buffers for Rust
• The Cargo Book: Specifying dependencies from git repositories - Who needs a registry anyway?
• cargo audit - Security vulnerability scanner for Rust dependencies
• epoll - Async I/O API in Linux
• Tokio - The async runtime powering Pingora
• mio - Tokio’s abstraction over epoll and other async I/O OS interfaces
• Noah Kennedy - An actual Tokio expert on the Pingora team
• Rain: Cancelling Async Rust - RustConf 2025 talk with many examples of pitfalls
• foundations - Cloudflare’s foundational crate for Rust project that exposes Tokio internal metrics
• io_uring - Shiny new kernel toy for async I/O
• ThePrimeTime: Cloudflare - Trie Hard - Big Savings On Cloud - “It’s not a millie, it’s not a billie, it’s a trillie”
• valuable - Invaluable crate for introspection of objects for logging and tracing
• bytes - Very foundational crate for reference counted byte buffers
• DashMap - Concurrent HashMap with as little lock contention as possible
• Prossimo - Initiative for memory safety in critical internet infrastructure
• River - Prossimo-funded reverse proxy based on Pingora
• Rustls - Memory-safe TLS implementation in Rust, also funded by Prossimo
• http crate - HTTP types for Rust
• h2 - HTTP/2 implementation in Rust
• hyper - Fast HTTP implementation for Rust
• ClickHouse Rust client - Official Rust client by Paul Loyd
• Pingap - Reverse proxy built on Pingora
• PR: Add Rustls to Pingora - by Harald Gutmann
• PR: Add s2n-tls to Pingora - by Bryan Gilbert

Official Links

• Cloudflare
• Cloudflare Blog
• Pingora on GitHub
• Edward Wang’s Blog Posts
• Kevin Guthrie’s Blog Posts

Building autonomous robots that operate safely in the real world is one of the most challenging engineering problems today. When those robots carry sharp blades and work around people, the margin for error is razor-thin.

In this episode, we talk to Andrew Tinka from Scythe Robotics about how they use Rust to build autonomous electric mowers for commercial landscaping. We discuss the unique challenges of robotics software, why Rust is an ideal choice for cutting-edge safety-critical systems, and what it takes to keep autonomous machines running smoothly in the field.
About Scythe Robotics
Scythe Robotics is building autonomous electric mowers for commercial landscaping. Their machines combine advanced sensors, computer vision, and sophisticated path planning to autonomously trim large outdoor spaces while ensuring safety around people and obstacles. By leveraging Rust throughout their software stack, Scythe achieves the reliability and safety guarantees required for autonomous systems breaking new ground in uncontrolled environments. The company is headquartered in Colorado and is reshaping how commercial properties are maintained.
About Andrew Tinka
Andrew is the Director of Software Engineering at Scythe Robotics, where he drives the development of autonomous systems that power their robotic mowers. He specializes in planning and control for large fleets of mobile robots, with over a decade of experience in multi-agent planning technologies that helped pave the way at Amazon Robotics. Andrew has cultivated deep expertise in building safety-critical software for real-world robotics applications and is passionate about using Rust to create reliable, performant systems. His work covers everything from low-level embedded systems to high-level planning algorithms.
Links From The Episode

• Ski trails rating - A difficulty rating system common in Colorado
• NVIDIA Jetson - Combined ARM CPU with a GPU for AI workloads at the heart of every Scythe robot
• The Rust Book: Variables and Mutability - Immutability is the default in Rust
• Jon Gjengset: Sguaba - A type safe spatial maths library
• The Rust Book: Inheritance as a Type System and as Code Sharing - Unlike Java, Rust doesn't have inheritance
• Using {..Default::default} when creating structs - The alternative is to initialize each field explicitly
• The Rust Book: Refutability - Rust tells you when you forgot something
• Clippy - Rust's official linter
• Deterministic fleet management for autonomous mobile robots using Rust - Andy Brinkmeyer from Arculus - 2024 Oxidize warehouse robot talk with deterministic testing
• ROS - The Robot Operating System
• Ractor - A good modern actor framework
• Rain: Cancelling Async Rust - RustConf 2025 talk with many examples of pitfalls

Official Links

• Scythe Robotics
• Scythe on LinkedIn
• Scythe on GitHub
• Andrew Tinka on LinkedIn

Are you one of over 240 million subscribers of Amazon's Prime Video service? If so, you might be surprised to learn that much of the infrastructure behind Prime Video is built using Rust. They use a single codebase for media players, game consoles, and tablets. In this episode, we sit down with Alexandru Ene, a Principal Engineer at Amazon, to discuss how Rust is used at Prime Video, the challenges they face in building a global streaming service, and the benefits of using Rust for their backend systems.

About Prime Video
Prime Video is a streaming service offered by Amazon that provides a wide range of movies, TV shows, and original content to its subscribers. With over 240 million subscribers worldwide, Prime Video is one of the largest streaming platforms in the world. In addition to its vast content library, Prime Video also offers features such as offline viewing, 4K streaming, and support for multiple devices. On the backend, Prime Video relies on a variety of technologies to deliver its content, including Rust, which is used for building high-performance and reliable systems that can handle the demands of a global audience.
About Alexandru Ene
Alexandru worked on the transition of Prime Video's user interface from JavaScript to Rust. He has been with Amazon for over 8 years and previously worked at companies like Ubisoft and EA. He has a background in computer science and is an active open source maintainer. Alexandru lives in London.
Links From The Episode

• Ferris Makes Emulators Ep.001 - The Journey Begins - First episode of a famous series where Jake Taylor wrote a Nintendo 64 emulator in Rust from scratch
• CMake - Very common build system used in C++ applications
• Conan - C++ Package Manager community project
• C++ Smart Pointers - Still a footgun
• Herb Sutter: The Free Lunch Is Over - The seminal 2005 paper that highlights the importance of concurrency, well past C++'s mainstream adoption
• Rust in Production: cURL - Baseline library used everywhere, written in C, but performant and safe
• Prime Video Platforms - One app runs on all of these
• WebAssembly (WASM) - Enabling Rust code with good performance that you can still download and run like JavaScript, avoiding the need for firmware updates on some devices
• Entity Component System - Used in the UI Rust code for pages in the app
• Bevy - Game engine written in Rust
• Leptos - UI framework that makes reactive programming in Rust easier
• tokio - The de facto standard async runtime for Rust
• SIMD - A nice feature set some CPUs support
• WebAssembly Micro Runtime - A tiny WASM runtime well suited for IoT platforms
• WebAssembly Working Group

Official Links

• Amazon Prime Video
• Rust & WASM for UI: Faster Prime Video on ANY Device - Alexandru Ene, QCon San Francisco 2024
• Alexandru Ene on LinkedIn
• Alexandru's Blog
• Alexandru Ene on GitHub

It’s time for another recap including our highlights of Season 4.

We’ve been at this for a while now (four seasons, and 32 episodes to be exact). We had guests from a wide range of industries: from Microsoft to Astral, and from password managers to satellite systems.

This time, it’s all about using Rust for foundational software, which is software that is critical to a team or even an entire organization. Rust is a great fit for this type of software!
We’ll take a break for the summer. In the meantime, we encourage you to catch up on any episodes you might have missed, and to check out the links below.

And if you’re interested in using Rust for your own projects, check out our dedicated learn page for additional content about Rust adoption or reach out to us directly for a chat.
Links From The Episode

• 10 Years of Stable Rust: An Infrastructure Story - Graydon Hoare’s reflection on Rust’s 10th anniversary
• Rust For Foundational Software - Matthias’ blog post on Rust’s role in foundational software

About Us

• Simon Brüggen (M3t0r) on GitHub
• Simon Brüggen (M3t0r) on Mastodon
• Matthias Endler on Mastodon
• About corrode Rust Consulting
• ‘Rust in Production’ Podcast on Apple
• ‘Rust in Production’ Podcast on Spotify

All Episodes of Season 4

If you missed any of the previous episodes, now is a great time to catch up:

• Episode 1: Microsoft with Victor Ciura
• Episode 2: Svix with Tom Hacohen
• Episode 3: Astral with Charlie Marsh
• Episode 4: Rust with Niko Matsakis
• Episode 5: Tembo with Adam Hendel
• Episode 6: 1Password with Andrew Burkhart
• Episode 7: KSAT with Vegard Sandengen

Credits
Podcast production by Simon Brüggen (M3t0r).
We would like to thank the guests for their time and insights. We would also like to thank you, the listener for your support and feedback. Hosting and producing a podcast is a lot of work, but it’s worth it when we hear from you. Here’s to another great season!

As a kid, I was always fascinated by space tech. That fascination has only grown as I've learned more about the engineering challenges involved in space exploration.

In this episode, we talk to Vegard Sandengen, a Rust engineer at KSAT, a company that provides ground station services for satellites. They use Rust to manage the data flow from hundreds of satellites, ensuring that data is received, processed, and stored efficiently. This data is then made available to customers around the world, enabling them to make informed decisions based on real-time satellite data.
We dive deep into the technical challenges of building reliable, high-performance systems that operate 24/7 to capture and process satellite data. Vegard shares insights into why Rust was chosen for these mission-critical systems, how they handle the massive scale of data processing, and the unique reliability requirements when dealing with space-based infrastructure.
From ground station automation to data pipeline optimization, this conversation explores how modern systems programming languages are enabling the next generation of space technology infrastructure.
About KSAT
KSAT, or Kongsberg Satellite Services, is a global leader in providing ground station services for satellites. The company slogan is "We Connect Space And Earth," and their mission-critical services are used by customers around the world to access satellite data for a wide range of applications, including weather monitoring, environmental research, and disaster response.
About Vegard Sandengen
Vegard Sandengen is a Rust engineer at KSAT, where he works on the company's data management systems. He has a Master's degree in computer science and has been working in the space industry for several years.At KSAT, Vegard focuses on building high-performance data processing pipelines that handle satellite telemetry and payload data from ground stations around the world. His work involves optimizing real-time data flows and ensuring system reliability for mission-critical space operations.
Links From The Episode

• SpaceX - Private space exploration company revolutionizing satellite launches
• CCSDS - Space data systems standardization body
• Ground Station
• Polar Orbit - Orbit with usually limited ground station visibility
• TrollSat - Remote Ground Station in Antarctica
• OpenStack - Build-your-own-cloud software stack
• RustConf 2024: K2 Space Lightning Talk - K2 Space's sponsored lightning talk, talking about 100% Rust based satellites
• K2 Space - Space company building satellites entirely in Rust
• Blue Origin - Space exploration company focused on reusable rockets
• Rocket Lab - Small satellite launch provider
• AWS Ground Station - Cloud-based satellite ground station service
• Strangler Pattern - A software design pattern to replace legacy applications step-by-step
• Rust by Example: New Type Idiom - Creating new wrapper types to leverage Rust's type system guarantees for correct code
• serde - Serialization and deserialization framework for Rust
• utoipa - OpenAPI specification generation from Rust code
• serde-json - The go-to solution for parsing JSON in Rust
• axum - Ergonomic web framework built on tokio and tower
• sqlx - Async SQL toolkit with compile-time checked queries
• rayon - Data parallelism library for Rust
• tokio - Asynchronous runtime for Rust applications
• tokio-console - Debugger for async Rust applications
• tracing - Application-level tracing framework for async-aware diagnostics
• W3C Trace Context - Standard for distributed tracing context propagation
• OpenTelemetry - Observability framework for distributed systems
• Honeycomb - Observability platform for complex distributed systems
• Azure Application Insights - Application performance monitoring service

Official Links

• KSAT
• Vegard on GitHub

Handling secrets is extremely hard. You have to keep them safe (obviously), while at the same time you need to integrate with a ton of different systems and always provide a great user-experience, because otherwise people will just find a way around your system. When talking to peers, a lot of people mention 1Password as a company that nailed this balance.
In today's episode, I talk to Andrew about how 1Password uses Rust to build critical systems that must never fail, how Rust helps them handle secrets for millions of users, and the lessons they learned when adopting Rust in their stack.
About 1Password
1Password is a password manager that helps users securely store and manage their passwords, credit card information, and other sensitive data. It provides a user-friendly interface and strong security features to protect users' secrets across multiple devices.
About Andrew Burkhart
Andrew is a Senior Rust Developer at 1Password in the Product Foundations org, on the Frameworks team and specifically on the Core Platform squad handling the asynchronous frameworks other developers use to build features (i.e. requests into the Rust core from the Native clients, data sync, etc.). He specifically specialized in that synchronization process, getting data federated from cloud to clients to native apps and back.
Links From The Episode

• Backend for Frontend Pattern - Architectural pattern for creating dedicated backends for specific frontends
• typeshare - Generate types for multiple languages from Rust code
• zeroizing-alloc - 1Password's minimal secure heap zero-on-free implementation for Rust
• arboard - Cross-platform clipboard manager written in Rust
• passkey-rs - Pure Rust implementation of the WebAuthn Passkey specification
• WebAssembly (WASM) - Binary instruction format for portable execution across platforms
• tokio - The de facto standard async runtime for Rust
• Clippy - A collection of lints to catch common mistakes in Rust
• cargo-deny - Cargo plugin for linting dependencies, licenses, and security advisories
• Nix - Purely functional package manager for reproducible builds
• Nix Flakes - Experimental feature for hermetic, reproducible Nix builds
• direnv - Load and unload environment variables based on current directory
• ACM: Spotify Guilds - A study into Spotify's Agile Model's Guilds
• axum - Ergonomic and modular web framework built on tokio and tower
• tower - Library for building robust networking clients and servers
• tracing - Application-level tracing framework for async-aware diagnostics
• rusqlite - Ergonomic wrapper for SQLite in Rust
• mockall - Powerful mock object library for Rust
• pretty_assertions - Better assertion macros with colored diff output
• neon - Library for writing native Node.js modules in Rust
• nom-supreme - Parser combinator additions and utilities for nom
• crane - Nix library for building Cargo projects
• Rust in Production: Zed - High-performance code editor built in Rust
• tokio-console - Debugger for async Rust programs using tokio
• Rust Atomics and Locks by Mara Bos - Free online book about low-level concurrency in Rust
• The Rust Programming Language (Brown University Edition) - Interactive version of the Rust Book with quizzes
• Rustlings - Small exercises to get you used to reading and writing Rust code

Official Links

• 1Password
• Andrew on GitHub
• Andrew on LinkedIn

Recently I was in need of a simple job queue for a Rust project. I already had Postgres in place and wondered if I could reuse it for this purpose. I found Tembo, a simple job queue written in Rust that uses Postgres as a backend. It fit the bill perfectly.

In today's episode, I talk to Adam Hendel, the founding engineer of Tembo, about their project, PGMQ, and how it came to be. We discuss the design decisions behind job queues, interfacing from Rust to Postgres, and the engineering decisions that went into building the extension.
It was delightful to hear that you could build all of this yourself, but that you would probably just waste your time doing so and would come up with the same design decisions as Adam and the team.
About Tembo
Tembo builds developer tools that help teams build and ship software faster. Their first product, PGMQ, was created to solve the problem of job queues in a simple and efficient way, leveraging the power of Postgres. They since made a pivot to focus on AI-driven code assistance, but PGMQ can be used independently and is available as an open-source project.
About Adam Hendel
Adam Hendel is the founding engineer at Tembo, where he has been instrumental in developing PGMQ and other tools like pg_vectorize. He has since moved on to work on his own startup, but remains involved with the PGMQ project.
Links From The Episode

• PostgreSQL - Super flexible ~40 year old relational database that just works
• R - Statistical Programming Language
• pgrx - Extend Postgres with Rust
• Postgres Docs: PL/pgSQL - Scripting with Procedural Language in PostgreSQL
• Postgres Docs: SPI - The Postgres Server Programming Interface
• pgmq - A lightweight message queue extension, initially written in Rust
• Tembo Blog: Introducing PGMQ - a blog post about the project
• sqlx - All of the great things of an ORM, without all of the bad things of an ORM
• tokio - The de facto standard async runtime for Rust
• AWS SQS - Amazon Web Services Simple Queue Service
• Postgres Docs: LISTEN - The native Postgres sub part of of pubsub
• Postgres Docs: NOTIFY - The native Postgres pub part of of pubsub
• tokio-stream - Tokio utility for asynchronous series of values
• Postgres Docs: Full Text Search - Postgres included FTS capabilities
• pgvector - The standard extension for vector/AI workloads in Postgres
• pg_vectorize - Automatically create embeddings for use with pgvector
• Python Standard Library: None - A type, but not an enum
• Rust in Production: Astral with Charlie Marsh - Massively improving Python day 1 experience
• Hugging Face candle - Use ML models in Rust

Official Links

• Tembo
• Adam on LinkedIn
• Adam on GitHub