Systems | Development | Analytics | API | Testing

In this episode of Data Builders Club, Henry Clavo shares lessons from over a decade in data engineering across healthcare and government, exploring what it really takes to build reliable data systems in the age of AI. From ETL best practices and data quality to AI hallucinations, observability, and the future of data engineering careers, this conversation is packed with practical insights for modern data teams.

Most data teams think MAR pricing is predictable and manageable. You pay for rows that change, your pipelines keep running, and the monthly bill should be easy to understand. But teams on the pricing model learn the reality 6 months in. If you’re on the MAR model this is a must watch (before you get your next bill spike). Dan Murphy explains why MAR mechanics are designed to break.

Somewhere along the way, data teams accepted a trade-off: pipelines that just work, or bills that you can actually forecast, pick one. So you live with the silent failures, the schema changes that break dashboards overnight, and the month-end invoice that never quite matches the data you moved. Not because it's acceptable, but because it's familiar. In this session, we're challenging that trade-off head-on. We'll break down where pipelines fail quietly and where costs inflate invisibly and show you, live, what it looks like when your pipeline gives you full visibility into every sync, every record, and every dollar.

The very system designed to keep you comfortable is the same one keeping you trapped. You pay only for rows that change. Simple, right? But the reality is quite different. MAR mechanics are designed to blow up when you scale. You expect to be paying for real data changes but you end up paying for the whole row, regardless of how little data you move on it. It lures you in with simplicity but the underlying mechanics is an alternate reality. Every connection follows its own pricing curve, the costs stop behaving logically, and forecasting your bill turns into a nightmare.

What does it really take to build AI-ready data systems people can trust? In this episode of Data Builders Club, James Serra shares lessons from 40+ years in data and AI, covering data quality, trust, Data Mesh trade-offs, real-time systems, and why strong foundations matter more than hype in the AI era. Featuring insights from James Serra, Data & AI Solution Architect at Microsoft and author of Deciphering Data Architectures.

Across 8 years and 2,000+ data teams in 40+ countries, three principles have shaped every decision we've made. That's the conviction behind Hevo's next chapter. In our latest video, Manish Jethani, Founder & CEO at Hevo Data, along with Scott Husband, Director of Partnerships, and Amit Gupta, VP of Engineering, walk through what's changed under the hood, and why every architectural decision traces back to three non-negotiables: Reliability, Simplicity, and Transparency.

AI is no longer an external layer. With Snowflake AI capabilities, combined with a real-time ingestion layer from Hevo, you can now build internal and external data products and AI workflows directly within Snowflake.

Everyone is investing in AI, but most teams are blocked by one thing: their data isn’t ready. Data is scattered across SaaS tools, pipelines break silently, and insights are delayed. Without fresh, reliable, and centralized data, AI models, dashboards, and real-time use cases simply don’t work.

Your data team doesn’t need more tools. It needs fewer bottlenecks. What if you could go from raw data to production-ready pipelines and AI workflows in a single day? With Snowflake’s Cortex Code, teams can now build, optimize, and deploy data workflows using natural language, dramatically accelerating development inside the warehouse.