Author Archives: Ben Lorica

Are Your AI Agents Flying Blind in Production?

Subscribe • Previous Issues Beyond Black Boxes: A Guide to Observability for Agentic AI The core mindset shift for agentic systems is simple: observability isn’t an add-on, it’s a production prerequisite. Enterprises are unwilling to trust black-box agents; they expect to understand behavior, decision-making, and reasoning. This means architecting for visibility from the very first design docContinue reading “Are Your AI Agents Flying Blind in Production?”

Multimodal lakehouses: The architecture AI teams are migrating to

When the “lakehouse” was first introduced in 2020, the goal was to reconcile data warehouses and data lakes in a single architecture: open formats on cheap object storage, with ACID transactions, schema enforcement, governance, BI support, and streaming built in. The promise was simple: one system for SQL analytics, data science, and machine learning, insteadContinue reading “Multimodal lakehouses: The architecture AI teams are migrating to”

The Rise of the Multimodal Lakehouse

Subscribe • Previous Issues Multimodal lakehouses: The architecture AI teams are migrating to When the “lakehouse” was first introduced in 2020, the goal was to reconcile data warehouses and data lakes in a single architecture: open formats on cheap object storage, with ACID transactions, schema enforcement, governance, BI support, and streaming built in. The promise was simple:Continue reading “The Rise of the Multimodal Lakehouse”

80% of AI Startups Are Built on Chinese Models: The New Reality for 2026

Chinese developers have overtaken American creators in open-model downloads, with partners at Andreessen Horowitz estimating that nearly 80% of new AI startups are now building on Chinese open-source infrastructure. This shift in developer mindshare stems from a vacuum in the US market: while both nations possess a comparable number of active labs, American releases haveContinue reading “80% of AI Startups Are Built on Chinese Models: The New Reality for 2026”

Legal AI Unpacked: What Works, What Fails, What’s Next

Law firms were early adopters of tools for searching and classifying large document collections, so their current enthusiasm for generative AI follows a familiar pattern. Business press coverage reinforces this interest: articles forecasting the disruption of knowledge work routinely single out legal services as particularly vulnerable to automation, given the profession’s reliance on text-intensive research,Continue reading “Legal AI Unpacked: What Works, What Fails, What’s Next”

AI’s biggest enterprise test case is here

Subscribe • Previous Issues Legal AI Unpacked: What Works, What Fails, What’s Next Law firms were early adopters of tools for searching and classifying large document collections, so their current enthusiasm for generative AI follows a familiar pattern. Business press coverage reinforces this interest: articles forecasting the disruption of knowledge work routinely single out legal services asContinue reading “AI’s biggest enterprise test case is here”

Gemini 3: Google’s Pitch vs. Users’ Reality

With each new foundation-model launch, the provider arrives with a familiar script: emphasize the novel capabilities, publish benchmark charts, and explain why this version is different from both its predecessors and its rivals. Users, for their part, increasingly assume that any new flagship model will proclaim state-of-the-art scores on an array of benchmarks. I’ve evenContinue reading “Gemini 3: Google’s Pitch vs. Users’ Reality”

Time Series Foundation Models: What You Need To Know

The recent emergence of Time Series Foundation Models (TSFMs) offers a powerful new tool for forecasting. Their effectiveness, however, is often constrained by an architectural design that analyzes each time series as an independent entity. This approach overlooks the rich, structured context available in most enterprise data warehouses, where a product’s sales history is interconnectedContinue reading “Time Series Foundation Models: What You Need To Know”

When Text Helps Time Series (and When It Doesn’t)

Subscribe • Previous Issues Time Series Foundation Models: What You Need To Know The recent emergence of Time Series Foundation Models (TSFMs) offers a powerful new tool for forecasting. Their effectiveness, however, is often constrained by an architectural design that analyzes each time series as an independent entity. This approach overlooks the rich, structured context available inContinue reading “When Text Helps Time Series (and When It Doesn’t)”