Cube

The future of financial reporting is not a chart. It’s an answer. In two seconds. We built Spaces to give finance teams an AI-powered workspace that turns questions into powerful insights, instantly. Learn more: https://bit.ly/4biYdRr

Three months in GA, and we're already transforming how teams like Brex and Drata build data workflows. We are the first analytics platform where AI builds your data model for you. Today, Cube agent already works in 200+ companies and has processed 500,000+ lines of semantic layer code. Every company wants AI-powered analytics. But there's a problem no one has solved: AI without a semantic layer hallucinates and gives you confident, wrong answers. The only fix is a strong data model that gives AI the context to understand your business. But building one by hand takes weeks. So the industry is stuck: build manually and wait, or let AI guess. We created a better option. Cube's AI writes the semantic layer itself.  The foundation that makes every query, dashboard, and answer accurate. No trade-off between speed and accuracy. Cube gives you both. Here's how it works: 1. Connect your data sources. Cube scans your schema and builds a semantic model in seconds. 2. Need a new metric? Describe it — AI writes the code on the fly. 3. Generate reports and dashboards with no hallucinations. Try Cube for Free: https://lnkd.in/e6YUzQEK

I have a bit of a quirky side habit: stress-testing AI agents. Most of them fail in predictable ways. They hallucinate tables and joins, infer weird semantics from schemas, and give plausible but incorrect answers. Frankly, a lot of what’s out there isn’t ready for real analytical work. I stress-tested Cube's new analytics agent, and it's one of the very few I'd actually trust today. Cube has been the OG semantic-layer and headless BI company for a long time, and that shows in how they approach agents. The key difference is the semantic layer. The agent queries semantic models, not raw schemas. That means it operates inside defined guardrails instead of improvising. In one test, I asked for data that didn’t exist, and it refused rather than hallucinating an answer. You get a pat on the back, AI. Disclosure: This video is sponsored by Cube. I had full editorial control, and these are my own tests and honest opinions.

Supercharging Ad Ops & Customer Success with AI + Cube At Cross Screen Media, we use Cube to power AI agents that help our Ad Operations and Customer Success teams with daily campaign adjustments and optimizations faster than ever. We’ve partnered with Cube since 2024, using its semantic layer as the foundation of our cross-screen measurement platform. Cube lets us define and dynamically compute complex metrics while using pre-aggregations to deliver consistently fast, scalable dashboards for our teams. In 2025, we started developing AI agents to optimize campaign workflows. Cube proved to be the perfect fit for this effort because it made our data easy to understand and simple for AI to access. The result for our internal teams has been dramatic: ✅ Tasks in campaign build-out and optimization that once took 15–30 minutes now take seconds. ✅ Our Customer Success and Ad Ops teams spend less time pulling data and more time delivering insights and results for advertisers. Big thanks to the Cube team for helping us move fast and bring AI into the daily workflows that matter most to our customers.

AI analysts are like any other data analyst. Giving them warehouse credentials and letting them run wild rarely ends in effective data analysis. It doesn't matter how intuitive your naming convention is for column and table names if they lack broader organizational context. When this happens, AI invents metric definitions that sound plausible but don't match how your business actually calculates revenue or churn. Different users get different answers to the same question. In our upcoming Deep Dive, we're showing how to fix this with Compass + Cube: → Why AI analytics fails without governance (and what that actually means in practice) → How semantic layers give AI the guardrails it needs to be trustworthy → Live demo: governed self-service analytics your data team can actually stand behind If you're exploring AI analytics but worried about accuracy and trust, this one's for you. Register today!