David Yaffe

Just released: You can deploy Estuary Flow’s powerful real-time data infrastructure in your private cloud. Secure data management, without compromises. Learn more: estuary.dev/private-deploy…

@EstuaryDev provides Change Data Capture (CDC) for many different databases, including Amazon's @dynamodb, @MongoDB, @PostgreSQL, @Oracle and more. Now you can integrate Estuary Flow directly with @startreedata for real-time analytics: startree.ai/resources/dyna…

@mattturck This is funnier than you got credit for @mattturck

DO YOU WANT TO GROW YOUR AUDIENCE ON X/TWITTER? Learn how this candidate MAXIMIZED his *new job announcement* and TRIPLED his follower count IN ONE DAY with just ONE simple hack - a 🧵 This morning Tonight

@mattturck So I guess you've deployed all of @FirstMarkCap's money @mattturck ?

My inbox: AI agents AI agents AI agents AI agents AI agents AI agents AI agents

@EstuaryDev @MaterializeInc Super excited for the partnership!

We've been cooking! Excited to showcase this tutorial on integrating Estuary Flow with @MaterializeInc! Set up a CDC flow for #sqlserver that aggregates incoming data using familiar SQL - in real-time with strict exactly-once delivery guarantees. estuary.dev/cdc-sqlserver-…

5. Philosophy - Franz Kafka believed in Existentialism.  People exist in a world where death is incumbent and he viewed the expectation for them to act morally as absurd. - Apache Kafka believes that datasets should have schemas….sometimes. That is all.

4. Topics - Franz Kafka mainly wrote about one topic, existentialism - Apache Kafka can handle writing to 100s of thousands of topics…depending on the number of partitions in each, but scaling those can get tricky.

People often ask me the difference between Franz Kafka and Apache Kafka... Explained below: 1. Storage - Franz Kafka has unlimited storage via libraries and home collections - Apache Kafka has a default 7 day retention

Today, we are excited to announce that we have raised over $11 million in funding, including a seed round led by @danielgross and @natfriedman , and a pre-seed round led by @mattturck at FirstMark, with participation from a number of industry leaders, including @tasso , Spencer Kimball, @calvinfo , Tristan Handy, Aston Motes, and many others. Our goal at Espresso is to accelerate compute, and we believe Gen AI is the path to get there. We've already improved the state-of-the-art in data warehouse optimization - our first product is reducing our customers' Snowflake bills by as much as 5x, automatically - but we're not stopping there. We plan to apply this technology to accelerate all of compute by 1000x. venturebeat.com/ai/espresso-ai…

@itsjoenaso Co-founder here and would be happy to connect if you are interested!

Ive heard good things about Estuary. Anyone have hands-on experience with it? Looking to revamp an ETL pipeline and trying to avoid Fivetran

If you're building Data Products without an immutable log, you're probably doing it wrong A data product log is an immutable append-only record of every change to a data product over time They enable time travel, flexibility building data products and interop between systems

How do you make #streamingETL and #ChangeDataCapture as simple as #BatchETL? Listen to the @DataStackShow as @dyaffe and @JGraettinger from @EstuaryDev explain the challenges, and their 10 year journey #ETL #dataengineering #streamingdata #ETLpipeline datastackshow.com/podcast/simpli…

@ben_brandwood @laurenbalik @fivetran It really depends on the use case, but all of those can be reasonable. For HFT, it means microseconds, but most people talk about it in terms of milliseconds!

This is excellent. Suing over termination for refusing to promote "real-time" data capabilities that aren't close to "real-time"... It is unethical to promote vaporware! @Fivetran promoting "real-time" anything should be on notice!

In the future, companies will mix and match batch and streaming seamlessly. Phil Fried explains how in this episode of Geek Narrator: youtube.com/watch?v=pOqQ-0…

Vector DB is a game changer in leveraging LLMs. But, vector DB alone for context isn't enough when using generative AI. Here’s one way to solve the context problem: estuary.dev/gpt-real-time-…

@laurenbalik @EstuaryDev @fivetran Love it Dr @laurenbalik

Dave and Johnny and co. at @EstuaryDev put out a great comparison and takedown of @fivetran pricing yesterday. 💪💪💪 estuary.dev/graduate-elt-t… This is vendor content 😱 but in my experience with using all of these tools it all tracks. Fivetran is just too overvalued, they have too many employees (1000+!) and GTM budgets to pay for, they don't deliver new connectors quickly because these may be unprofitable for their Monthly Active Row model, and they've been resting on their laurels of marketing and overcharging customers and hoping customers don't notice/care. Well, people notice and care. The hip, new thing is optionality. You should be able to turn knobs around cost and latency around the stack on vendor costs, ingest, transform. If you have hundreds or thousands of dbt models and you're using Fivetran, ask yourself why that is! Many times it's because you're moving the complexity right to dbt and many of the extra dbt models are the result of the fixed schemas Fivetran makes. You're just delaying addressing complexity and you're doing it on a more expensive and often slower cycle! Also, being able to turn the knobs on latency is critical. Fivetran only offers 60 minute, 15 minute, 5 minute levels but those are made up, who is to say those are right? Maybe you want real-time or maybe you want 20 minutes. This upsell of latency that Fivetran enforces grinds my gears more than anything. If you - the customer! - are doing WAL or you have many sources with varying rate limits on APIs, you should have complete control over turning your knobs on when and how you ingest. 💪💪💪 Dr. Lauren gives Estuary a stamp of approval for many use cases. It can be a great solution. ----- We're up to 19 Fivetran screenings so far on my crusade 💪 Keep 'em coming if you want to get off Fivetran and cut the nonsense. There are many ways to do better with better vendors, better performance, and fewer billings games. Book time for your personal Fivetran screening below -- or book time just to say high or for any therapy! calendly.com/lauren-upright…

@5theschaton @laurenbalik I'm a co-founder of @EstuaryDev @5theschaton and would be happy to connect if you'd like!

@laurenbalik Hey Lauren! We’ve been using fivetran to fill in cloud sources that segment doesn’t get, but now I’m thinking we should be doing estuary + hevo /keboola but I’m not 100% sure. Is this the kind of thing these sessions are for discussing?

I'm excited to announce that I've opened up Fivetran Health Screenings more formally this week based on demand. (And I'm sticking with the medical motif!) 💉😷😂💊 There is no reason for businesses to keep forking over money and creating critical pipeline dependencies on Fivetran. Book a time below if you'd like to have a free, candid session to go over how to get your organization off Fivetran - or just say hi! calendly.com/lauren-upright…