Factor House (@factorhousehq) header
Factor House (@factorhousehq) profile picture

The control plane for Kafka, Flink, and Iceberg.

Melbourne, Australia
Joined September 2019
Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)Photo by Factor House (@factorhousehq)
How does a team of ~20 run Kafka for one of North America’s largest banks? 20+ clusters, 300+ initiatives, 20B+ events on peak days. Thanks to TD’s Julia Lok and Sandy Yang for sharing it at #Current2026 with our CEO Derek Troy-West.
38
We’re at Databricks Data + AI World Tour in Chicago on Oct 29 🇺🇸 Streaming Kafka or Flink data into the lakehouse? Come find the Factor House team at booth #6
1
8
Heading to Current in San Francisco? We’re giving 3 talks Nov 4-5: OpenLineage, Kafka ops from the terminal, and multi-tenant Kafka with TD Bank. Come say hi
12
Move MirrorMaker 2 onto Kafka Connect the obvious way and you'll re-replicate every record from offset zero. Our guide covers seeding offsets, why offset translation is off by default, and a cutover order that keeps it uneventful. Link in comments.
1
1
17
Kafka DLQs often mean adding a stream processor or retry service. But you can handle retries and DLQs consumer-side using Kafka alone, with per-consumer-group topics so failed messages only replay to the group that failed. No extra infrastructure. Link in the thread.
1
12
Three nines of availability still allows almost 9 hours of downtime a year. Chad Harris (Factor House) & Justin George (NetApp Instaclustr) break down what SLA tiers really cost when comparing Confluent Cloud, MSK, and open source Kafka.
1
1
34
Thanks to everyone who joined last night's Melbourne Apache Kafka meetup at Factor House. Karel Sague shared practical techniques for cutting Kafka costs, and Dominique Ronde from Confluent talked Kafka, Flink and GenAI. Great turnout, great questions.
1
78
New in Kpow: Signals. Automated insight into your Kafka cluster health, right when you need it. Flags misconfigurations, partition skew, unbalanced brokers, and idle consumer groups before they become incidents. Tracked as a health score over time. Demo below.
1
1
111
We are happy to announce that every release of Kpow for Apache Kafka and Flex for Apache Flink will now come with a published VPAT (Voluntary Product Accessibility Template). We are here for engineers. 🚀 factorhouse.io/blog/articles…
1
68
Kpow for Apache #Kafka is WCAG 2.1 AA Compliant. Learn what Web Accessibility means to Factor House: factorhouse.io/blog/articles…
86
Lean how to find the needle in a #Kafka haystack with our guide to searching topics with Kpow and kJQ. factorhouse.io/blog/how-to/q…
72