writing/
Long-form writing on building AI and data products inside regulated firms. Mostly agents, mostly governance, mostly the hard parts.
Of Models & Harnesses
The code builders wrote to steer models is coming apart in three directions: down into the models, open to the agent running inside it, and out to the institution that answers for what it does. What that means for regulated firms, and what I expect by 2027.
Of Models & Harnesses
Welcome to Enterprise AI Theater
So you think you're doing AI? The gap between what executives believe and what anyone can measure, why the chatbot era ended, and how to plan for what replaced it.
Welcome to Enterprise AI Theater
AFIX — An Open Schema for Cross-Enterprise Agent Governance
From Profile to Interchange In my first article in the AI governance series, I compared the AI governance evolution to the FIX moment. It ended ended with a specific proposal: build a Financial Services Profile for Google’s A2A protocol: domain-specific extensions declaring agent identity…
AFIX — An Open Schema for Cross-Enterprise Agent Governance
Building the Governance Layer: A Practitioner’s Guide to Intra-Enterprise Agentic Governance
Your agent deleted 200 files from a shared drive last Tuesday. It had permission. The files matched the pattern. But some were active deal documents. Who authorized “matches the pattern” as sufficient criteria for a destructive action, and is that authorization written down anywhere? Your eval…
Building the Governance Layer: A Practitioner’s Guide to Intra-Enterprise Agentic Governance
Is Agentic AI Approaching Its FIX Protocol Moment?
You retrained your model last week. How do you know it’s better than the one it replaced? Not “the benchmarks improved”, but: how do you certify it’s better, with the kind of documented evidence that survives a regulatory audit? Your research agent just pulled a client’s portfolio data to generate…
Is Agentic AI Approaching Its FIX Protocol Moment?
Claude Code’s in-built JIRA
Claude Code just shipped a built-in task management system with dependency tracking. Yep, it’s basically JIRA! No, you don’t have to manage it yourself. Claude handles task creation, status updates, owners and blocking dependencies automatically. But there’s a catch: This feature is in its infancy…
Claude Code’s in-built JIRA
Of Models & Harnesses
The code builders wrote to steer models is coming apart in three directions: down into the models, open to the agent running inside it, and out to the institution that answers for what it does. What that means for regulated firms, and what I expect by 2027.
Welcome to Enterprise AI Theater
So you think you're doing AI? The gap between what executives believe and what anyone can measure, why the chatbot era ended, and how to plan for what replaced it.
AFIX — An Open Schema for Cross-Enterprise Agent Governance
From Profile to Interchange In my first article in the AI governance series, I compared the AI governance evolution to the FIX moment. It ended ended with a specific proposal: build a Financial Services Profile for Google’s A2A protocol: domain-specific extensions declaring agent identity…
Building the Governance Layer: A Practitioner’s Guide to Intra-Enterprise Agentic Governance
Your agent deleted 200 files from a shared drive last Tuesday. It had permission. The files matched the pattern. But some were active deal documents. Who authorized “matches the pattern” as sufficient criteria for a destructive action, and is that authorization written down anywhere? Your eval…
Is Agentic AI Approaching Its FIX Protocol Moment?
You retrained your model last week. How do you know it’s better than the one it replaced? Not “the benchmarks improved”, but: how do you certify it’s better, with the kind of documented evidence that survives a regulatory audit? Your research agent just pulled a client’s portfolio data to generate…
Claude Code’s in-built JIRA
Claude Code just shipped a built-in task management system with dependency tracking. Yep, it’s basically JIRA! No, you don’t have to manage it yourself. Claude handles task creation, status updates, owners and blocking dependencies automatically. But there’s a catch: This feature is in its infancy…
Parallel Paths: Finding Validation for My Context Engineering Framework
Back in September 2025, I realized the hard way that the traditional way of using Prompt Engineering for agentic tasks, while suitable for early frontier models, was not going to be suitable or scalable for complex tasks. But as time has gone by, LLMs have only gotten smarter and AI agents are now…
From Context Chaos to “CLEAR” – An Origin Story
The No-Code Journeyman I started AI-assisted prototyping with no-code tools like Lovable, Bolt, and Figma Make, but hit frustrating limitations: platform lock-in, no code downloads, and hidden implementation details. Debugging was painful without transparency into changes, and I couldn’t prevent or…
AI-Assisted Rapid Prototyping – Part IV
In Part IV of this AI-assisted Rapid Prototyping series, I’ll focus on the fourth step: Realize & Crystallize. As a reminder, my prototyping process has five stages, outlined below. The Realize and Crystallize stages are inseparable; this is where you actually build & iterate on the prototype…
AI-Assisted Rapid Prototyping – Part III
Welcome to Part III of this series in which I will focus on the Strategize phase of rapid application prototyping. If you haven’t already, you can Step 2: Strategize The Strategize phase focuses on developing the master prompt for Claude and Gemini, our chosen assistants. Within this initial…
AI-Assisted Rapid Prototyping – Part II
Welcome to Part II of this series in which I will focus on the thought process I use before starting to prototype an application from zero into something tangible. If you haven’t already, you can read Part I here: Prototyping Applications using Gemini CLI & Claude Code Together – Part I. As…
Prototyping using Gemini & Claude – Part I
It is truly an exciting time to be in Product right now. The world of AI has dropped a whole new set of cheat codes into our laps, giving us incredible ways to accelerate our workflows and boost productivity. For me, the magic wasn’t just in automating tasks; it was in reclaiming my calendar. By […]
nothing matches.
try broadening the search or clearing tag filters.