San Francisco, United States of America
51 - 200
Information Tech
2021
Rust, C++, Go, JavaScript, TypeScript, Amazon Web Services (AWS), Kubernetes, Docker, Grafana
checkDiverse team
checkContinuous feedback
checkCommunicating transparently
checkValues teamwork
checkFlexible hours
checkOpen source contributor
checkResults-driven
checkFast-paced environment
checkQuality & Innovation
Neon is building open-source cloud-native PostgreSQL. Our architecture separates storage from compute, allowing for stateless and serverless Postgres. We're a well funded startup with deep knowledge of Postgres internals and decades of experience building databases. We are a systems company; we work on low-level code with strict performance and correctness requirements.
Neon was created by a team of Postgres hackers and led by CEO Nikita Shamgunov (co-founder of SingleStore). Neon is built on open-source principles and is focused on giving back to the Postgres and developer communities.
Neon is a distributed team of 90 people working from Europe, North America, and the Middle East. We are built on principles of the open-source Postgres community: transparency, contribution, accountability, kindness, and professional ethics.
Opportunity
Benefits
Join us on a visual tour through our offsite gathering in Barcelona, explore the best moments, thrilling activities, and bonding experiences.
Every frame showcases our culture, serving as a testament to our collaborative spirit that propels us toward greater heights.
Discover the heart of our team away from the office and get a glimpse into our vibrant corporate culture.
Neon is now Generally Available. We've shipped major improvements to Neon internals that, combined with our operating experience scaling up to 700,000+ databases over the past year, give us the confidence that Neon is ready to support your business-critical workloads. If you're building or scaling an application, you like Postgres, and you prioritize efficiency and engineering velocity, Neon is for you. Sign up for free.
Postgres continues to hold its position as one of the most popular developer databases ever created. While MySQL lost some of its luster after its acquisition by Oracle, Postgres continues to win more developer trust. In 2023, it topped the poll in the StackOverflow Developers Survey and in the same year, was voted the DBMS of the Year by DBEngines. The current internet meme is “Just Use Postgres” for everything.
Despite its popularity, Postgres suffers from the two problems that beset nearly all relational databases. It is difficult to scale up and down without interruption. Worse, it takes a long time to restore production operations when software bugs affect the data.
This happens because with Postgres it's hard to create a high-fidelity developer environment as it requires copying the full state of the database. With that, maintaining multiple copies of data and schema for every developer is a major pain in the neck. The database becomes the nexus of development for most teams and invariably is the bottleneck constricting developer velocity.
When the Neon engineering team approached these issues they knew they had to solve several problems with today’s database development lifecycle:
Almost every app needs a database, and while developer tools and workflows evolved with platforms like GitHub and Vercel, databases did not evolve to match the developer experience of these systems. Evolving developer experience around the database requires a fundamental rethinking of the database architecture. But this is not a “look at our cool new database” engine. We know developers love and want to use authentic open-source Postgres. So, at Neon, we looked at how we could change the experience of developing on Postgres without losing everything that makes Postgres great.
We borrowed an idea from Amazon Aurora, the separation of storage and compute; and we aren’t talking about simply using a network-attached disk like EBS. True separation of storage and compute opened up the possibility to scale both parts of the service independently. Unlike AWS Aurora we decided to open source all the changes in Postgres and also send them upstream as well as fully open source our cloud native storage.
We made architectural decisions that ensured that our users have the complete Postgres experience, including Extensions, internals, command line and management tools. Everything that worked on-premise should work in our cloud service.
An additional benefit from separating storage and compute is – well we have a brand new cloud native storage implementation and we can “do things” in it. Controlling storage allows us to implement branching – a flagship feature that unlocks a better developer experience for building database apps.
This means we can create branches at any point in time and start a compute instance pointing to that database version. So, every developer can work on a different branch without colliding with their colleague's work.
Compute can now scale independently of storage so the CPU size and count can accurately match the load. Even large database loads have peaks and troughs. We wanted to avoid developers having to size for the largest expected load.
A database branch is copy-on-write, so it only points to a set of already created and stored pages. So, branches take less than a second to create. Because they are copied on write branches incur no space overhead until data is written to them.
Suppose a developer fat fingers a table or a database out of existence? No problem, move the branch to the second before that event happened. Starting a new database or recovering from a production failure happens in seconds because the Neon architecture doesn’t require you to ship hundreds of gigabytes of data to create new instances of databases and servers.
Users don’t want databases, they want data. Neon is designed to take the overhead of starting, stopping, creating, and cloning databases from hours to seconds. The major velocity inhibitor of modern development is the relational database. Imagine if data was easy, what would it do to your developer productivity?
With Neon database branching is integrated directly into the developer workflow and without interruption. When database data and/or schema changes take hours to propagate across the development team everyone suffers.
Neon is the next big pivot for Postgres, and all of these changes have been offered upstream to the Postgres community. They are available in our public repo today. Our mission is to provide everything Postgres developers need and less. Less waiting for clusters to start, less waiting for databases to restore, less waiting for colleagues changes to deploy, less waiting for the DBA team to schedule their updates.
During our Preview we’ve been focused on building a solid technical foundation for Neon. Some of our favorite milestones:
Neon Postgres is Generally Available. Serverless autoscaling and instant branching are built for modern developers who have to iterate fast. Go faster with Neon, it’s ready, give it a spin.
We don’t have any open positions right now.