Apurva Mehta

Apurva Mehta

@apurvamehta.com

Co-Founder and CEO of @responsive.dev. I write about stream processing and company building. Views are my own. 🌐 apurvamehta.com

2.1K followers80 following185 posts

Apurva Mehta reposted

Apurva Mehta reposted

SlateDB 0.5.0 is out! Features: - Checkpoints - Clones - Read only client - Split/merge database foundation - TTL filtering on reads - Last version with breaking byte format changes By the numbers: - 62 commits - 2 new contributors - 10 total contributors github.com/slatedb/slat...

github.comRelease v0.5.0 · slatedb/slatedbWhat's Changed Refactor Block Tests to Use Table-Driven Test Cases by @samsond in #410 Update await calls in README.md by @criccomini in #425 chore: Apply table driven test for sst.rs by @jeffreyl...

Apurva Mehta reposted

Apurva Mehta reposted

I’m excited to listen to this one. Some of my favorite folks all in one podcast 😀

Tech on the Rocks@totrrocks.bsky.social

New episode: Reinventing Stream Processing with @apurvamehta.com from @responsive.dev We cover: Real-time vs low latency in streaming Why stateful streaming is hard at scale RocksDB, Postgres limitations SQL vs flexible APIs Decoupling state & compute with SlateDB. 🎧 👉

This week, our 101 series has @ableegoldman.bsky.social diving deep into the world of performance optimization for Kafka Streams. This isn't just theory: she shows how we realized 10x throughput improvements by implementing these techniques in prod! www.responsive.dev/blog/perform...

responsive.devKafka Streams 101: Optimizing your apps for maximum performanceLearn the concepts, configs, and application design strategies to get your Kafka Streams apps running as optimally as possible.

Apurva Mehta reposted

New episode: Reinventing Stream Processing with @apurvamehta.com from @responsive.dev We cover: Real-time vs low latency in streaming Why stateful streaming is hard at scale RocksDB, Postgres limitations SQL vs flexible APIs Decoupling state & compute with SlateDB. 🎧 👉

techontherocks.showTech on the RocksJoin Kostas and Nitay as they speak with amazingly smart people who are building the next generation of technology, from hardware to cloud compute. Tech on the Rocks is for people who are curious…

Apurva Mehta reposted

We're keeping the ball rolling on the Kafka Streams 101 series with yet another blog post! This time the focus is on configuration, and breaking down the most important configs to focus on for various needs (such as correctness, resiliency, etc) Check it out here: www.responsive.dev/blog/importa...

responsive.devKafka Streams 101: The Most Important ConfigsLearn which configurations are most critical for a reliable and efficient deployment

Apurva Mehta reposted

We are happy to announce that we have completed another SOC2 Type 2 audit along with completing another successful penetration test against our cloud services. You can find the latest reports on our trust center: trust.responsive.dev

trust.responsive.devResponsive Trust CenterResponsive is a stream processing solution that brings disaggregated state, autoscaling, observability and enterprise-grade support to Kafka Streams. Eliminate unpredictable lag and rebalance instabil...

(1/3) One of the hardest problems in the world of stream processing is upgrading stateful applications. Why? Because the events never stop, and your application has to handle the events correctly before and after a upgrade and if rewind your position in a topic.

Apurva Mehta reposted

Had two people ask me for guidance about open source foundations this week. Lots of trepidation around Apache, not much familiarity with CNCF, and curiousity about why I chose Commonhaus for SlateDB. Here are my thoughts:

cnr.shComparing Apache, CNCF, and Commonhaus | cnr.shI've used open source projects for over 30 years and contributed for about 20 of those. My first interaction with an open source foundation was with Apache when I began working with Apache Hadoop ...

This will be the foundation for being able to branch @kafkastreams.bsky.social apps to support seamless blue/green deploys. It also allows you to branch a previous version of the state to debug an issue from the past. Powerful stuff coming to the world of stream processing!

Chris@chris.blue

SlateDB now has clones 🤯 Users can clone an existing DB's data to a new location. It's nearly instantaneous since it references the data from the old bucket rather than copying. Writes to the clone update the new location. Compaction lazily merges old data into the new directory./ht @responsive.dev

I'm enjoying putting the finishing touches on my next blog post on the Why, When, and How of Kafka applications. It's a distillation of my point of view from a decade in the space and I hope it's as useful to read as it was fun to write. Coming Tuesday!

I've used ChatGPT and/or Gemini to brainstorm ideas for presenting concepts visually multiple times recently. They have unlocked ideas I wouldn't have thought of otherwise. And sometimes, they generate the graphics. For the rest, Figma is perfect. It's such a major boost either way!

The first of our Kafka Streams 101 Series. Practical knowledge on thinking about windowing and time in @kafkastreams.bsky.social We are going to ship new tutorials regularly over the next few months, and you can find them all here: www.responsive.dev/resources/ka...

Almog@almog.xyz

Kafka Streams 101: Windows & Time! 🕰️ What's the difference between event, stream and wall clock time? 🪟 What are the four different types of windows? ⁉️ What are the important error messages and metrics, and what do they mean? 👉 Read the full lesson here: www.responsive.dev/blog/windows...

Apurva Mehta reposted

There are some learnings that only come from having seen a wildly successful technology go from a nascent project to ubiquitous. These insights can change the trajectory for new tech. Jason's summary about error handling is a great read for practical engineers: github.com/slatedb/slat...

github.comAPI Error Guidance · Issue #454 · slatedb/slatedbDumping a few thoughts on API errors. At the moment, we have a single flat error SlateDBError which is exposed through the public API. This is the same pattern used in Kafka and there are some less...

Sophie arguably has the best @kafkastreams.bsky.social blog posts on the internet. Now you get a say on what she should write about next!

Sophie Blee-Goldman@ableegoldman.bsky.social

[1/5] Hey fellow Kafka Streams otters! You may have already seen my previous blog posts such as the "Debugging Rebalances" and "EOS guide" from last year (and if you haven't, go check them out! Links in thread) But now it's 2025, and time for me to think about what to write next...

👋

Kafka Streams@kafkastreams.bsky.social

Kafka Streams has officially come to Bluesky! Follow this account for updates on everything Kafka Streams and keep up to date with new features, bug fixes, and upcoming KIPs. Long live the otter!

It was so much fun to observe the @responsive.dev team brainstorm distributed state snapshots for Kafka Streams using SlateDB/RS3. Nothing beats the thrill of having a team of the best engineers solving hard problems with nothing but a whiteboard in front of them!

Great to see so many people online. If you want to hear about how Metronome scaled their Kafka streams apps to 10's of TBs of state and 10's of thousands of events/s, do tune in for the last few minutes!

Almog@almog.xyz

Starting in 10 minutes! If you are in production with Kafka Streams or curious about how Metronome handles insane throughput in realtime, you won't want to miss this at 9:30AM PT. www.linkedin.com/events/q-a-s...

Apurva Mehta reposted

Apurva Mehta reposted

For any Kafka Streams fans out there who want to know what it's like scaling an EOS application up to multiple TB of state at ~10k+ records/s with minimal downtime -- make sure to catch the Q&A we're hosting tomorrow with Metronome. Join us Dec 12th at 9:30am PST! www.linkedin.com/events/72647...

linkedin.comQ&A: Scaling Event Driven Architectures with Metronome | LinkedInWe're hosting a live Q&A session with two engineers who have seen it all: - Casey Crites, a founding engineer at Metronome, implemented and scaled the Kafka pipeline at Metronome from zero to where i...

The best thing about doing a startup is that you have the absolute freedom to invest in creative ideas with absolute accountability. The bad thing is that 99% of your ideas will fail. The most important asset in my experience is being able to use failure as a fuel for more creativity.

It's remarkable how popular Kafka Streams is. Here's data for people searching for solutions to their Kafka Streams problems and landing on @responsive.dev. That's an up-and-to-the-right chart I love! On that note, what's the best book or blog you've read for optimizing website conversions?

This was a fun conversation with @thegeeknarrator.bsky.social and @yingjunwu.bsky.social about the world of SQL stream processing on one side and event driven apps on the other. Thanks for having me on the show, Kaivalya!

Kaivalya Apte@thegeeknarrator.bsky.social

Excited to announce the next episode on The GeekNarrator podcast on Stream Processing. Really great insights from two different worlds of stream processing, @responsive.dev and @risingwave.bsky.social

Apurva Mehta reposted

"SQL is a great language for expressing ETL and analytics logic, but it has never been a language developers pick for expressing application logic. By contrast, Responsive was founded on the belief that the true value of stream processing is in the new applications you can build with event streams."

responsive.devFounder's Letter: Responsive and the Future of Stream ProcessingA recap of all that we've built in our first 18 months as a company, and what that means for the future of stream processing.

Apurva Mehta reposted

KIP-1104 is completed. It closes such an obvious gap in the FK-join API that I am wondering why it was not done this way in the first place... (and why did it not come up much earlier?) Well, we will have it with upcoming 4.0 release. Better now than never. cwiki.apache.org/confluence/d...

cwiki.apache.orgKIP-1104: Allow Foreign Key Extraction from Both Key and Value in KTable Joins - Apache Kafka - Apache Software Foundation

Is stream processing interesting because of the new apps it enables, or because it promises better data processing? I'm in the first camp and am proud of the major contributions Responsive has made to the application space. My reflections on what's next: www.responsive.dev/blog/respons...

responsive.devFounder's Letter: Responsive and the Future of Stream ProcessingA recap of all that we've built in our first 18 months as a company, and what that means for the future of stream processing.

With ChatGPT, higher level decision making will increasingly become bottleneck for any given project, and so its existence should really favor startups. For example, here's how I used ChatGPT during the rebuild of my personal website. (1/3)

View full thread

I'm no expert in the modern frontend stack, but with ChatGPT I could produce something useful pretty fast. And the benefits will compound from here. So a startup can stay small and still produce world class output without needing to hire as much for special skills. (3/3)

With ChatGPT, higher level decision making will increasingly become bottleneck for any given project, and so its existence should really favor startups. For example, here's how I used ChatGPT during the rebuild of my personal website. (1/3)

As an uber calculator: it just solved specific problems, and was at least directionally right all the time. As a tutor, it was an amazing guide through novel solution spaces (2/3)

I've always wanted to have a landing page for apurvamehta.com. But a blog was too much of a commitment and a short bio was too simple. @bsky.app allows you to display your feed on a static webpage, which is a perfect setup and is perfectly aligned with the 'your domain is your id' philosophy too!