Bangalore Apache Kafka®, Data & AI

4.5•915 ratings

About us

This group is for people interested in Apache Kafka, stream processing, Applied AI and ecosystem. If you are interested in presenting at a future event, please fill out the following form https://forms.gle/KmtmZZYn1TVJMNAk9

Upcoming events

See all

IN PERSON! Apache Kafka® Meetup Bangalore- Aug 2026
Sat, Aug 29 · 11:00 AM IST
Zopdev, ZopSmart, 24th Main Rd, 22nd Cross Rd, Parangi Palaya, Sector 2, Bengaluru, Karnataka 560102, Bangalore, IN
Hello everyone! Join us for an IN PERSON Apache Kafka® meetup on Aug 29 from 11:00AM, hosted by ZopDev in Bangalore!

***
IMPORTANT!
Please fill out this form for expense and reporting purposes, as it is mandatory for entry.
***

📍 Venue:
409, 24th Main Rd, 22nd Cross Rd, Parangi Palaya, Sector 2, Bengaluru, Karnataka 560102, India

***
Agenda:
- 11:00 - 11:10: Welcome
- 11:10 - 11:50: Shabeeb R P, Staff Software Engineer I, Confluent
- 11:50 - 12:30: Vishal M, Software Engineer, Datazip & Ankit Sharma - Lead Engineer, Datazip
- 12:30 - 12:40: Break
- 12:40 - 13:20: Avinash, Zopsmart
- 13:20 - 14:30: Lunch
***
💡 Speaker:
Shabeeb R P, Staff Software Engineer I, Confluent

Talk:
Iceberg V3 and beyond

Abstract:
Iceberg has evolved quite a bit from V1 to V3, with each version bringing new capabilities to the table. In this talk, we’ll take a quick look at how Iceberg has evolved across V1, V2, V3, and what’s coming in V4. We’ll spend most of the time on V3, covering features like Variant data type, Deletion Vectors, Row Lineage, Default Values, and more, along with the problems they are designed to solve. We’ll also look at how these features change the way we work with and manage Iceberg tables. Finally, we’ll wrap up with a brief look at the V4 spec and what’s coming next.

-----
💡 Speaker:
Vishal M, Software Engineer, Datazip
Ankit Sharma - Lead Engineer, Datazip

Talk:
Making Kafka Ingestion Scalable & Reliable: Rebalancing, Exactly-Once, and Upserts

Abstract:
Kafka is widely used as a source for data ingestion, but long-running ingestion jobs introduce unique challenges around consumer group rebalancing, offset management, exactly-once processing, and upserts. In OLake, syncs can run for hours or even days, making a rebalance particularly risky: partitions can be reassigned mid-sync, leading to reprocessing, duplicates, or inconsistent state.
This talk explores why these challenges become harder for long-running ingestion workloads, how OLake approaches reliable rebalance handling and exactly-once processing, and how it will support upserts when ingesting Kafka data.

-----
💡 Speaker:
Avinash, Zopsmart

Talk:
TBC

Abstract:
TBC
135 attendees