You can now stream OpenTelemetry logs, traces, and metrics directly into Tinybird data sources. Deploy ClickHouse compatible schemas to production with one command.

ClickHouse Grafana plugin compatibility

You can now use the ClickHouse Plugin for Grafana to connect to and explore data (OpenTelemetry or otherwise) from your Tinybird workspaces.

Announcement blog post|Docs|Demo Video

25th June

Wednesday

Windows Support and Onboarding Improvements

Want to build a real-time analytics agent? Find resources and examples below.

26th June

Thursday

Sinks in Forward

Export the results of your Tinybird pipes to Kafka, S3, or GCS -- on a schedule or on demand. Now available in Forward.

26th June

Sinks in Forward

Kafka Sink

Export pipes results to Apache Kafka or any Kafka-compatible consumer.

S3 Sink

Export pipes results to Amazon S3 -- with support for multiple file formats, file partitioning, and compression.

GCS Sink

Export pipes results to Google Cloud Storage -- with support for multiple file formats, file partitioning, and compression.

Announcement blog post|Docs|Demo Video

27th June

Friday

Performance and Infrastructure Improvements

We tackle infrastructure complexity so you don't have to. Here are 6 big performance + infra upgrades we just shipped.

27th June

Performance and Infrastructure Improvements

We ditched Prometheus for autoscaling (and don't miss it)

Tinybird uses KEDA and its own real-time analytics platform to autoscale Kafka workloads. Learn how we made it work.

We cut AWS costs by 20% while scaling faster with Karpenter

Read how Tinybird slashed AWS costs by 20% —and up to 90% on CI— with Karpenter, Spot Instances, and real-time autoscaling. A deep dive into our setup, lessons learned, and tips you can steal.

Scaling up ClickHouse ingestion with a Multi-writer architecture

Sometimes compute-compute separation isn't enough. Here's how we horizontally scaled our ClickHouse streaming ingestion service to handle bigger loads more efficiently.

How we protect your primary workloads with Compute-Compute Separation for Populates

With compute-compute separation for populates, Tinybird now handles even the most demanding data transformations and backfills without impacting your primary workloads or having to temporarily scale your cluster for more compute resources.

How we made our ingestion pipeline 30% faster with C++ (not Rust, sorry)

JSON to ClickHouse RowBinary conversion powers much of Tinybird's ingestion. We made it more efficient, complete, and predictable, cutting CPU usage by more than 30%.

How we automatically handle ClickHouse schema migrations

How we use data lineage to optimize ClickHouse table deployments and avoid unnecessary and expensive data migrations.

Launch Week

tinybird.ai

Tinybird MCP Server

Birdwatcher Analytics Agent Template

Birdwatcher Slack App

Explorations UI

Tinybird OpenTelemetry Exporter

Tinybird OpenTelemetry Exporter

ClickHouse Grafana plugin compatibility

Windows Support and Onboarding Improvements

Windows support for Tinybird CLI

Onboarding flow improvements

Sinks in Forward

Kafka Sink

S3 Sink

GCS Sink

Performance and Infrastructure Improvements

We ditched Prometheus for autoscaling (and don't miss it)

We cut AWS costs by 20% while scaling faster with Karpenter

Scaling up ClickHouse ingestion with a Multi-writer architecture

How we protect your primary workloads with Compute-Compute Separation for Populates

How we made our ingestion pipeline 30% faster with C++ (not Rust, sorry)

How we automatically handle ClickHouse schema migrations

Subscribe to hear first about our daily Launch Week releases.