Airbyte

Wear these costumes to scare the heck out of data engineers: 🎃 Ghost Pipeline (hasn't synced since 2022) 💀 Zombie Connector (deprecated API, still in prod) 🧛 Vampire Query (drains compute at 3 AM) 🧟 Frankenstein's Data Stack (15 tools, zero docs) 🕷️ Spider Schema (nested JSON nightmare) Happy Halloween from team Airbyte!

Our October edition of Data Bytes is here! It’s full of new updates, including details on Airbyte Plus for growing teams, major performance improvements to our Snowflake destination, Airbyte support for Apache Polaris catalog integration, and more! Dive in, and subscribe today to never miss an edition!

At TechCrunch Disrupt, Airbyte CEO Michel Tricot led a roundtable discussion tackling the biggest barrier to AI adoption that most founders don't see coming: data infrastructure. For startups with lean teams, building AI-ready infrastructure can feel impossible. Especially when you're trying to ship fast. Michel's roundtable covered the practical strategies that actually work: ⚡️ How to build data pipelines that don't break the bank or your team ⚡️ Managing both structured and unstructured data for AI contexts ⚡️ Securing secrets and permissions without slowing down development ⚡️ Ensuring agents can act responsibly with your data If you're building with AI, your infrastructure needs to be ready before your model is. The startups that get this right early are the ones that scale. 📈

We've been listening 👂 Based on conversations with our customers, we saw a need for a new pricing plan for Airbyte. Today, we're introducing Airbyte Plus—the same powerful Standard plan you know, now with annual billing and accelerated support. What changed: ✅ Fixed annual pricing (no monthly surprises) ✅ Accelerated support for faster resolution times What stayed the same: → All 600+ pre-built connectors → Best-in-class CDC methods for databases → Schema propagation and column selection → Connector Builder for custom integrations → Full API access, Terraform provider, and PyAirbyte → Integrations with Airflow, Dagster, and Prefect Airbyte Plus is designed for growing teams (20-300 employees) who need production-grade data integration with predictable costs and premium support. We launched Airbyte Plus because you asked for it. Now let's build something great together. https://lnkd.in/gik8m2Nf

Large companies have increased productivity 5.5% since the release of ChatGPT. Small companies? Down 12.3% 📉 It’s not a lack of investment. Intuit estimates 68% of small businesses have integrated AI into their operations. Here’s what I think is driving the gap: data infrastructure 👷 I've worked with many small companies on AI initiatives. I see the challenges they face, and understand why large enterprises have a huge advantage when it comes to delivering value with AI. They have clean, accessible first-party data and a unified data context layer feeding models. They’ve established Infrastructure that connects legacy systems to agentic workflows. Small companies? They’re still working out how to de-silo their data. We are not talking about tech startups, but rather the majority of traditional small businesses that fuel the global economy. You can't build useful agentic applications when half your data is trapped in disconnected systems or your team is manually stitching together CSVs. This gap will continue to widen unless small businesses invest in the data that fuels their AI. The good news? It’s never too late to bet on your data foundation. 

CodeSandbox needed flexibility. 🧘 As their platform grew with thousands of daily active users, they needed a data integration solution that could adapt to evolving product needs and new use cases. Their challenge: Find product-market fit fast while managing thousands of daily active users and multiple data sources. The problem with their old setup: ❌ Meltano's UI wasn't intuitive ❌ Stitch struggled with large Postgres databases ❌ Both tools slowed down their ability to iterate quickly Enter Airbyte. ✅ Seamless integration of new data sources into their architecture ✅ Granular control over data syncing through flexible APIs ✅ Open-source transparency for understanding and adapting integrations ✅ Freedom to build custom connectors as their needs evolve Read the full story: https://bit.ly/3Scw5VL

Happening 🔜 Small Data SF We'll be there with a booth both days. Stop by to talk about Airbyte 2.0, data sovereignty, and how we can help when your data doesn't need to take a trip to the cloud. Think small, develop locally, ship joyfully. 🚀

I bought my first computer with money I earned picking tomatoes as a kid 🍅 Funny how those early experiences shape you. The repetitive work, the small wins adding up to a big reward. These lessons helped me when I was a founder building something from scratch. At Airbyte, we're doing a lot of unglamorous work behind the scenes. Building connectors, fixing edge cases, making data pipelines just... work. But this time, I’m not doing it alone. I love it. And having a team that celebrates the small wins along the way? That makes the work worth it!!!

❄️ We just wrapped our Snowflake destination overhaul, and honestly, the numbers are wild!!! Snowflake is one of our largest destinations by both volume and active connections, so we knew optimizing it would matter. What we didn't expect is that the results would be this good. What we shipped: ⚡ Up to 10x faster syncs  💰 95%+ cost reduction We crushed storage costs, and slashed compute. This is the kind of impact that changes how teams think about their data infrastructure: faster pipelines, lower bills, same reliability. If you're running Snowflake at scale, this one's for you 🚀