Web Finds on Data & AI From Around the Net

Explore our collection of web resources picked from across the internet to empower your skills and knowledge. Reading from the Keboola team which you could also find interesting.

Check Out the Stuff We Like

• Review of Data Orchestration Landscape — An independent review comparing modern workflow orchestrators (Airflow, Prefect, Dagster, Mage), with candid insights on each tool's strengths, weaknesses, and ideal use cases.
• Orchestration Showdown: Dagster vs Prefect vs Airflow — Breaks down the differences between three leading data pipeline orchestrators, highlighting each platform's unique features and guiding teams on which might best fit their needs.
• What Makes a Great Consultant? — Some engineers may not like this one.
• 2024 Data Orchestration Pricing Deep Dive — Examines how major orchestration platforms price their offerings (from Airflow to Prefect and Dagster), revealing striking differences in cost models that data teams should factor in when budgeting for pipeline tools.
• Why Data Product Management May Be Doomed — Argues that the buzz around data products could turn into a fad because many organizations adopt the term without implementing true product management practices or delivering real user value.
• DuckLake. We love DuckDB. We are eager to see what's next.
• Data Product Management – How to Think About This Function — Clarifies the responsibilities and value of a Data Product Manager in large organizations, explaining how this often-misunderstood role can be defined and aligned with business needs to drive better outcomes.
• Implementing Data Product Management Framework — Outlines a step-by-step framework for treating data as a strategic product — from clear ownership and governance to agile development processes — to help organizations maximize the value and impact of their data assets.
• How CEOs Are Using Gen AI for Strategic Planning — Shows how executives are leveraging generative AI to brainstorm ideas and identify opportunities in strategy development, while cautioning that these tools must augment (not replace) human judgment given their limitations in foresight and firm-specific knowledge.
• State of AI: Global Survey (2024) — Highlights a global survey's findings on enterprise AI adoption — more companies report revenue boosts and workflow redesigns from generative AI, yet only about 1% consider their AI initiatives fully mature, underscoring that most firms are still early in their AI journey.
• 2025 AI Business Predictions — Predicts that top-performing companies will integrate AI deeply into their core strategy (moving beyond ad-hoc use cases), and foresees developments like AI copilot agents amplifying workforce productivity and a growing emphasis on Responsible AI as key to sustaining ROI.
• The High Cost of Misaligned Business and Analytics Goals — Explains how companies can waste time and money when analytics projects aren't aligned with business strategy, emphasizing the need for a strong data culture, executive buy-in, and clear goals to truly convert data investments into business value.
• Data vs. Business Strategy – Which Is Responsible for What? — Provides a pragmatic look at becoming data-driven by delineating the roles of data strategy versus business strategy, clearing up common misconceptions so that data initiatives directly support and inform the company's overall strategic goals.
• Where Data-Driven Decision-Making Can Go Wrong — Details common pitfalls that leaders face when relying on data (from misinterpreting statistics to overlooking context) and offers guidance on avoiding these traps so organizations can make better decisions grounded in sound analysis.
• Inside the Snowflake–Databricks Rivalry, and Why Both Fear Microsoft — Reveals the fierce competition between Snowflake and Databricks — from secret project codenames to cheeky billboard ads — and explains why both rising data giants regard Microsoft as a formidable threat in the battle for cloud data platform dominance.
• Modern Data Stack: An Expensive Mess — Argues that the fashionable modern data stack has turned into an overly complex and costly tangle of tools and cloud services, cautioning that without consolidation or better management, its promised benefits can be undermined by runaway integration costs and operational chaos.
• The Rise and Fall of the Vector Database Category — Discusses how the boom in vector databases (spurred by AI's need to store and search embeddings) is now leveling off as traditional databases and search engines add vector capabilities — illustrating that new features don't always require a whole new infrastructure category.