Feldera
@feldera.bsky.social
Built on DBSP theory (Best Paper, VLDB ‘23), Feldera recomputes precisely what changes, even in 40K-line SQL pipelines with thousands of joins, window functions, and recursion, with the exact same correctness guarantees as your warehouse.
There’s a fascinating algorithm called DBSP that single-handedly managed to combine streaming and batch systems and solve a decades-old database efficiency problem. Lalith Suresh joined me to explain the core of how it works: 😊 youtu.be/CyvnH8OUCUA
Making Materialized Views Actually Fast with DBSP (with Lalith Suresh)
YouTube video by Developer Voices
youtu.be
How do customers convert 10,000+ lines of production Spark SQL into running Feldera pipelines in minutes? They hand it to their AI agents powered by our latest skill, Felderize.
Once you've experienced a True Incremental View Maintenance (IVM) engine, recomputing everything from scratch sounds absolutely absurd. Too slow and too expensive to justify. We set up a little match of our own: Databricks Incrementally Materialized Views vs. Feldera.
We just improved Feldera support for deletion vectors and column mapping in Delta Lake tables. What does this mean for you? You can now experience True Incremental View Maintenance without having to move your data.
80% of enterprise Feldera pipelines are now authored by agents. So we built them skills.
Watch what happens when we run a simple 1v1 comparison against ClickHouse. As we turn up the volume on the transaction records, things start to really cook. Just not for us. Feldera barely broke a sweat. Incremental view maintenance never degrades at scale.
Right now, somewhere, a data engineer is getting another message from stakeholders: “Why is the Gold layer late?” It is a familiar reminder that the traditional medallion architecture was not built for real-time freshness. It is being held back by legacy batch processing.
A true incremental compute engine's latency should be *flat* no matter how much your data grows (O(delta) vs O(size of data)). Let's see this in action.
3 seconds versus 226 seconds can mean you’re flagging fraud or writing it off as a loss.
Our co-founder and Chief Scientist Mihai Budiu is speaking at @hasgeek Rootconf’s special conference on Databases on Friday, June 12 at 9:40PM PT virtually or Saturday, June 13 at 10:10AM IST Bangalore.
⚡️ Shipped This Week ✨ New profiler UI: Our pipeline profiler UI got a major glow up. Now in one view see your dataflow graph, your SQL, and a tabbed panel of live metrics, logs, and triage suggestions. When something needs triage, everything you need to assess the situation is in one clean view.
Databricks Data + AI Summit is the one place where everyone in the room knows the medallion architecture intimately, and nobody talks about what it actually costs you to keep it fresh.
Just because your data keeps growing doesn’t mean your compute bill needs to. Batch is inherently wasteful, recomputing everything on every run. The result: your cloud bill scales with the size of your data, even if only a few rows changed.
May newsletter is out! Fun fact: we just shipped our 300th release. 🎉 We're celebrating by shipping even more great stuff. Take a look. www.linkedin.com/pulse/may-ed...
May Edition 2026
🎉 We just hit our 300th release! We were founded in May 2023 with the deep belief that rerunning batch jobs from scratch when 99.9% of the data is unchanged was a fundamentally broken model.
linkedin.com
How do you give AI agents live, actionable signals for dynamic access control? Swipe to find out 👉
AI agents are only as good as the information they see, and only as fast as they can access it.
⚡️ Shipped This Week The teams rebuilding their data infrastructure aren't waiting for a better time. Every week this engine ships what they need, and the list keeps growing.
⚡️ Shipped This Week We keep tuning our engine to be faster and easier to observe. A couple of highlights:
⚡️ Shipped This Week We ship big and we ship small. This week was about the details. The kind that makes the engine easier to debug, observe, and run efficiently in production. A few highlights from this week:
April was a month of momentum. The team shipped across every layer of the stack: core engine performance, connector resilience, more SQL functions, and the OSS community showed up in a big way. Read the full newsletter here: www.linkedin.com/pulse/april-...
April Edition 2026
April was a month of momentum. The team shipped across every layer of the stack: core engine performance, connector resilience, more SQL functions, and the OSS community showed up in a big way.
linkedin.com
A great contribution from our OSS community! @flak153 shipped a Postgres CDC input connector for Feldera. Built with crash-safe replication, snapshot and streaming support, and fault tolerance baked in. Postgres pipelines can now read historical data and live changes in a single connector.
⚡️Shipped This Week More SQL functions. Pipelines got more observable. Memory usage went down. And the Feldera community keeps showing up. Here are some highlights from this week:
Feldera is SOC 2 Type 2 compliant ✅ As a bring-your-own-cloud platform, your data always stays in your infrastructure. SOC 2 Type 2 is our commitment to holding ourselves to the highest security standards. Independently audited by @PrescientSecurity. Full details: trust.feldera.com
⚡️Shipped This Week Every week we move fast and we build in the open. This week the Feldera team and a community contributor shipped features that make your pipelines more powerful, more resilient, and easier to operate at scale. Here are some highlights from this week:
Your agents are ready to make decisions in real time. Is your data platform ready to provide live data? Feldera is. Try it now: github.com/feldera/feld... Tell us about your experience with agentic real-time decision-making.
How do you give AI agents live, actionable signals? Stay tuned. 👀