lakeFS Acquires DVC, Uniting Data Version Control Pioneers to Accelerate AI-Ready Data
The lakeFS Control Plane for AI-ready Data provides agents that rely on large, multimodal datasets, isolated access, verifiable results and built-in governance. TL;DR A new kind of consumer for your data A few weeks ago at TrailblazerDX 2026, Salesforce put a name on something the rest of the industry had been circling for months: Headless […]
Unless you’ve been living under a rock, you’ve probably heard of multimodal data and its integration, now a standard feature of modern data platforms. As systems ingest data ranging from structured tables to unstructured text, graphics, and streams, the difficulty shifts from data collection to data integration. What differentiates experimental pipelines from production-grade systems is
AI projects often end up failing due to data, not models. Inconsistent inputs, poor data quality, a lack of lineage, and fragmented workflows subtly weaken even the most sophisticated algorithms. As datasets grow in size and complexity, data management for AI projects evolves from a supporting role to a core engineering discipline. Without a solid
Scaling AI isn’t about building better models; it’s about building the system around them. Without consistency in data, workflows and governance, teams hit the same walls: A Center of Excellence (CoE) for Enterprise AI solves this by standardizing how AI is built, validated, and deployed – so teams can move faster without losing control. But
Building infrastructure for AI-ready data is a technical challenge, that’s for sure. But it’s also a strategic imperative for organizations looking to scale AI across the board. As datasets grow and models become more complex, teams will look for platforms that can consistently store, analyze, manage, and disseminate data throughout the AI lifecycle. Without a
Modern data systems are subject to constant change – new data arrives, pipelines evolve, and tables get updated many times a day. Without the ability to check the state of data at any point in the past, teams are forced to work in the dark when troubleshooting issues. On top of that, reproducing results becomes
As AI adoption evolves and teams advance from scattered ML trial projects to running AI as a production system, they inevitably face the question of how to operate such a system reliably at scale. This is where an AI Center of Excellence (AI CoE) comes in. It’s an organizational and technical response to that shift:
Modern data operations call for more than just lightning-fast queries and scalable storage. Safety, reproducibility, and control are all key parts of the equation. As Iceberg becomes the foundation for analytical and AI-driven applications, how you handle changes to your tables will determine whether your data platform is resilient or unstable. Iceberg branching is a
You probably heard the saying “Garbage in, garbage out.” It holds true for any data system but properly prepared and handled data is especially critical to the success of AI deployments. A study from IBM Institute for Business Value (IBV) showed that just 16% of AI programs are reaching enterprise level. How can teams create,
The AI agent revolution is here. Coding agents like Claude Code, Cursor, and Codex are writing production software. Infrastructure agents are provisioning cloud resources. Data agents are transforming pipelines and updating dashboards. The trajectory is clear: agents will do more, not less. But there’s a pattern hiding in plain sight, one that explains why coding
AI systems increasingly influence decisions made in financial services, healthcare, and public services. This means teams need to be able to demonstrate that their models are transparent, fair, secure, and reproducible. Compliance is no longer just about policies and documentation – it requires technical guarantees embedded directly into data pipelines and ML workflows. Modern AI
Based on my presentation at PyData Global 2025. My colleague Yoav recently wrote about why reproducibility matters so much in healthcare AI and how data version control addresses the gap. This post is a follow up to that, with an overview of how to incorporate data version control into your existing ML workflow to address
We use cookies to improve your experience and understand how our site is used.
Join our community of experts:
introduce yourself, share your knowledge and discover best practices from fellow peers
