Arc's First Month: 358 Stars, Production POCs, and Small Data SF
Ok, we made it. I can't believe how many things happened in the last 30 days since we released Arc to the public as a technical preview (or alpha, whatever you want to call it).
First, we want to say THANK YOU for the engagement, feedback, and for trying Arc. Without all of that, our last 30 days wouldn't have been the same.
Let's review what happened in the last month
On October 6, 2025, we made the first commit and opened Arc Core to the public, licensed under AGPL, with one objective in mind:
- Understand if there's hunger for a time-series database that truly separates storage from compute, that is part of the modern data stack using DuckDB as a Query Engine.
I think the theory was proven. What we accomplished in the last 30 days—from the people running Arc, to the POC we're already running with a healthcare sector prospect, to the noise we made among some competitors—ufff. I think we're on the right track.
So, let's review what we did in the last month:
- 140 commits adding new features, fixing a lot of bugs, and reverting architectural decisions
- 358 stars on GitHub (thank you!)
- 18 forks
- 64 Arc instances running out there—most are Docker deployments, but we're seeing native installations too. We don't track use cases or locations (privacy first!), just instance counts
- POC ongoing with a company manufacturing medical devices in the US. They're evaluating Arc to replace their InfluxDB deployment for real-time clinical patient monitoring
- Replaced InfluxDB in our own monitoring stack—we're now using Arc to ingest metrics and logs
- Benchmarked Arc and found it's the fastest time-series database on c6a.4xlarge, the standard size for ClickBench
- Launched docs.basekick.net
- Released a Superset integration with support for querying via https://github.com/Basekick-Labs/arc-superset-dialect and https://github.com/Basekick-Labs/arc-superset-arrow endpoints
- Released a https://marketplace.visualstudio.com/items?itemName=basekick-labs.arc-db-manager to interact with Arc
- Released a https://github.com/Basekick-Labs/grafana-arc-datasource (Data Source) that we're using and looking for sign-off to complete the distribution
- Working on a https://github.com/influxdata/telegraf/pull/17875 for Arc that we've already pushed and are fine-tuning based on feedback (thanks to the InfluxData team!)
- Traveled to San Francisco for the Small Data conference (I'm still in SFO as I write this)
Here's what we shared at Small Data SF:
What surprised us
A ton of things happened, but here's what really caught us off guard:
The ecosystem integration hunger - We expected people to want a time-series database. What we didn't expect was how quickly the community asked for integrations. Within days, we had requests for Grafana, Superset, Telegraf, and more. The ecosystem matters more than we thought.
Reverting architectural decisions - Those 140 commits? Many were us saying "nope, that was wrong" and going back. We initially tried to be too clever with some Python optimizations. Turns out, simpler is faster. DuckDB does the heavy lifting—we just need to get out of its way.
People are running Arc in production already - This one keeps me up at night (in a good way?). We're still in technical preview, but some folks are already using Arc for real workloads. No pressure, right?
A ton of things, but what's next?
How the next 30 days look
We want to keep talking with you to understand what you need and how we can help solve pain points and challenges in sectors like Energy, Oil and Gas, and Finance—covering use cases around IoT, Telemetry, Observability, and more.
We also want to keep working toward releasing the first binary of Arc in December with release 2025.12.1. Binaries will be available for Linux, Mac, Windows, Docker, and Kubernetes, so the distribution work ahead is quite interesting.
We want to keep refining what we have and start planning the next thing to release next year, which is...
Arc Cloud
Think Arc, but managed and globally distributed. Storage-compute separation taken to the cloud. We don't want to give too much away just yet, but stay tuned to this blog—we'll be sharing more details soon.
One more thing
We need your help. We need your feedback. Talk to us. We want to build things that solve real problems for developers, engineers, and businesses—and we want to do it with an open heart, where vendor lock-in isn't a thing, where we listen to our users and serve them.
You can start testing Arc here
github.comBasekick-Labs/archttps://github.com/Basekick-Labs/arc
Thank you for everything.
Until next time.