At Rerun, we're building a new multimodal data stack for physical AI. Large parts of the physical-world economy will be transformed by these technologies, and we're building the data infrastructure and tools that enable it to happen. Data volumes grow big quickly with these applications and performance matters everywhere; from micro interactions to high throughput jobs.
With multimodal data stack we mean tools for extraction, ingestion, storage, querying, streaming, and visualization of temporal multimodal data. Rerun's open source project is already loved by some of the best teams in the world, and focuses on visualizing and handling multimodal logs. You'll join a newly formed team building out Rerun's commercial offering for ingesting, indexing and querying multimodal data at scale; a purpose-built analytical database that's already in private preview with signed customers and running at petabyte scale.
What you'll doYou will work on a cloud-native analytical database system built for multimodal (often time-series) data.
The system is built in Rust and makes extensive use of Apache Arrow. It supports specialized queries to handle robotics-style multimodal logs (very sparse and wide tables) and domain specific operators (e.g. related to 3D transforms).
Note: prior robotics knowledge is not necessary
You will help the system scale as usage grows, evolving the architecture, distributing work across nodes and systems, and tackling the bottlenecks that appear as load and data volumes increase.
You will design, implement and operate the core pieces that make this data-intensive system highly scalable, reliable and observable.
3-5 years with Rust, or a strong background in another systems language (C/C++) and real motivation to pick up Rust
Experience at a managed database product, data-platform, or high-traffic system operating at scale
Experience building backend systems in production, with a growing appetite for design and architecture ownership
Experience scaling a system as demand grows; evolving the architecture to handle much more load or data, not just making one component faster
Familiarity with distributed systems and their tradeoffs (coordination, consistency, failure modes, scaling across nodes)
Comfort with the layers between your code and the hardware; you don't need to be a database-internals specialist, but you enjoy digging in
Familiarity with concepts like filtering, indexing, or query execution
Good communication in English; you'll be in a close-knit remote team
Exposure to Apache Arrow or Apache Datafusion
Cloud deployment experience (AWS/GCP/Azure), even in a supporting role
Interest in machine learning, robotics, AR/VR, or simulation
Experience with GPU workloads
This role is open to candidates based in Europe.
We're a remote company headquartered in Stockholm, Sweden, with a growing hub in San Francisco.
We meet up in person for a week roughly once a quarter.
We're spread across European and US time zones. Rather than one fixed window, each team keeps a few hours of daily overlap that fits where its members are — for this team, that's a few hours of overlap in the afternoon CET.
We've put together an uncommonly talented tech team, value agency and helpfulness highly, and expect everyone to take broad responsibility for what they build.
We offer competitive cash and equity compensation, six weeks paid vacation, and whatever hardware and software you need to do your job.
Parental leave: 12 weeks full pay for every parent, alongside your local statutory leave entitlements.
Most of all we hope you're excited to build some really cool technology that will have a real impact on the world with us!
If you find that you don't fit neatly into this or any of the other open roles, we're not that rigid and are generally looking for great developers who can contribute to what we're building. Just go ahead and apply to the one you think is the closest and we'll take it from there!
