The following article includes clips from a recent Live with Tim O’Reilly interview. You can watch the full version on the O’Reilly Media learning platform. Addy Osmani is one of my favorite people to talk with about the state of software engineering with AI. He spent 14 years leading Chrome’s developer experience team at Google,…
Think about what happens when you onboard a new employee. First, you provision them tools. Email access. Slack. CRM. Office software. Project management software. Development environment. Connecting a person to the system they’ll need to do their job. However, this is necessary but not sufficient. Nobody becomes effective just because they can log into Salesforce….
This post first appeared on Addy Osmani’s Elevate Substack newsletter and is being republished here with the author’s permission. TL;DR: Aim for a clear spec covering just enough nuance (this may include structure, style, testing, boundaries. . .) to guide the AI without overwhelming it. Break large tasks into smaller ones versus keeping everything in one large…
Agentic AI has clearly moved beyond buzzword status. McKinsey’s November 2025 survey shows that 62% of organizations are already experimenting with AI agents, and the top performers are pushing them into core workflows in the name of efficiency, growth, and innovation. However, this is also where things can get uncomfortable. Everyone in the field knows…
For most of the past decade, AI governance lived comfortably outside the systems it was meant to regulate. Policies were written. Reviews were conducted. Models were approved. Audits happened after the fact. As long as AI behaved like a tool—producing predictions or recommendations on demand—that separation mostly worked. That assumption is breaking down. As AI…
Most multi-agent AI systems fail expensively before they fail quietly. The pattern is familiar to anyone who’s debugged one: Agent A completes a subtask and moves on. Agent B, with no visibility into A’s work, reexecutes the same operation with slightly different parameters. Agent C receives inconsistent results from both and confabulates a reconciliation. The…
My first post made the case for what a semantic layer can bring to the modern enterprise: a single source of truth accessible to everyone who needs it—BI teams in Tableau and Power BI, Excel-loving analysts, application integrations via API, and the AI agents now proliferating across organizations—all pulling from the same governed, performant metric…
I’ve been telling myself and anyone who will listen that Agent Skills point toward a new kind of future AI + human knowledge economy. It’s not just Skills, of course. It’s also things like Jesse Vincent’s Superpowers and Anthropic’s recently introduced Plugins for Claude Cowork. If you haven’t encountered these yet, keep reading. It should…
In a previous article, we outlined why GPUs have become the architectural control point for enterprise AI. When accelerator capacity becomes the governing constraint, the cloud’s most comforting assumption—that you can scale on demand without thinking too far ahead—stops being true. That shift has an immediate operational consequence: Capacity planning is back. Not the old…
On February 28, OpenAI announced it had reached a deal that will allow the US military to use its technologies in classified settings. CEO Sam Altman said the negotiations, which the company began pursuing only after the Pentagon’s public reprimand of Anthropic, were “definitely rushed.” In its announcements, OpenAI took great pains to say that…