Articles
Field notes
Everything I’ve published from the seat — 69 field notes on running engineering when agents do the executing and I stay accountable for the outcome.
Latest
Start here
Agentic Overwatch: Why Your Next Dev Team Will Look Like a NASA Control Room
·3102 words·15 mins
Agents don’t just write code anymore. They run ops, security, QA, data, and support, around the clock, while we still govern them with a team …
I'm Pro-AI. That's Exactly Why I'm Worried About Our Next Senior Engineers
·1124 words·6 mins
A guide for engineering managers on growing junior developers in an AI-heavy world, and for junior developers who want to stand out beyond just being …
AI Makes Code Cheap to Produce. Not Cheap to Own.
·700 words·4 mins
AI accounts for 42% of committed code. 96% of developers don’t fully trust the output. Only 48% always verify before committing. The gap between …
Stop Reviewing Code. Start Reviewing Evidence.
Agent-heavy teams ship twice the PRs at more than twice the size, and reviews wait almost five times longer for pickup. The review model built for …
The Shift
Execution moved to agents. The work — and who owns it — did not.
Developer Work Did Not Change. The Sequence Did.
·1508 words·8 mins
AI doesn’t make the job different. It changes when parts of the job happen, turning Monday morning from archaeology into editing.
When AI Writes 90% of Your Code, What Are You Actually Doing?
·1405 words·7 mins
Anthropic’s CEO says Claude writes 90% of code for most teams. If you think that means developers are obsolete, you’ve missed the point …
Org Charts for AI Agents: Mapping Your Human and AI Workforce
·3193 words·15 mins
AI agents aren’t coming. They’re already here, doing real work, while most organizations are still debating how to use ChatGPT. If …
Alberta Scanned 466 Million Lines of Code in 20 Hours. The Architecture Is the Story.
A Canadian province ran Claude Code across 466 million lines of government code in 20 hours — work they priced at 6.5 years the old way. The number …
The IDE Is Becoming Mission Control
·837 words·4 mins
Cursor 3 rebuilt its UI around agents. GitHub calls Agent HQ ‘mission control.’ VS Code is ‘your home for multi-agent …
The Cost
Cheap to produce is not cheap to own. The bill arrives downstream.
AI Makes Code Cheap to Produce. Not Cheap to Own.
·700 words·4 mins
AI accounts for 42% of committed code. 96% of developers don’t fully trust the output. Only 48% always verify before committing. The gap between …
AI Didn't Replace Software Engineering. It Made Bad Engineering Easier to Ship.
·1596 words·8 mins
The culture shifted. ‘Ship fast with AI’ became the expectation. Anyone who slows down to think looks unproductive. Discipline became a …
The Vibe Coding Backlash Is Right. Seniors Are Losing the Argument Anyway.
·1323 words·7 mins
Forbes just said vibe coding will break your company. Senior engineers are organizing against it. The data is on their side: independent audits keep …
Make the Cheap Path the Default. Make the Expensive Path Prove It.
Half of most AI agent bills is waste with a good story: context pollution, verification theater, tool schemas nobody reads, retries nobody capped. …
SaaS Is Dead. We Just Haven't Stopped Paying for It Yet.
·1966 words·10 mins
The bargain that powered SaaS for two decades was simple: rent generic software because custom is too expensive. That bargain is breaking. The cost of …
Judgment
The scarce asset — and how engineers grow it when AI does the typing.
I'm Pro-AI. That's Exactly Why I'm Worried About Our Next Senior Engineers
·1124 words·6 mins
A guide for engineering managers on growing junior developers in an AI-heavy world, and for junior developers who want to stand out beyond just being …
What's Holding You Back from Succeeding in the AI Era?
·3041 words·15 mins
I’ve watched teams double their output with AI, and I’ve also seen developers stall and managers struggle. The difference isn’t the …
The End of Courses: Learn From AI Like a Toddler, Or Become Obsolete
·2517 words·12 mins
Remember when shipping an app meant 40 hours of video courses and weeks of syntax memorization? An agent builds it in three minutes now. The 40-hour …
Overwatch
Supervision as architecture: control rooms, evidence gates, agents on the org chart.
Agentic Overwatch: Why Your Next Dev Team Will Look Like a NASA Control Room
·3102 words·15 mins
Agents don’t just write code anymore. They run ops, security, QA, data, and support, around the clock, while we still govern them with a team …
Stop Reviewing Code. Start Reviewing Evidence.
Agent-heavy teams ship twice the PRs at more than twice the size, and reviews wait almost five times longer for pickup. The review model built for …
Everyone Trained Their Engineers on AI. The Gap Didn't Move.
82% of organizations now offer AI training. 59% still report a skills gap. Security spent twenty years and 19,500 employees learning why: knowledge …
The One-Man Show Company. Don't Let the Monkeys Touch Production.
·2833 words·14 mins
A company used to start with people. Now it can start with one person and a swarm of AI agents that draft, build, test, sell, and support faster than …
The Context Engine: What Comes After We've Solved Code Generation
·827 words·4 mins
We’ve largely solved the context problem in AI coding tools. RAG systems ingest our docs, codebase-aware assistants understand our architecture, …
The Perimeter
Agents are an insider workforce. Secure them like one.
Prompt Injection 2.0: The New Frontier of AI Attacks
·1631 words·8 mins
Prompt injection has evolved from toy demos to sophisticated attacks targeting production AI systems. What was once a curiosity is now a genuine …
Building AI Systems That Don't Break Under Attack
·1805 words·9 mins
Understanding the threats is step one. Building defensive architectures that actually work in production is step two. Here’s what’s …
Securing the AI Supply Chain: The Threat Nobody's Talking About
·2088 words·10 mins
We’re building AI systems on top of models, datasets, and tools we don’t control. The supply chain is the attack vector nobody’s …
AI Security Isn't a Tool Problem, It's a Culture Problem
·2121 words·10 mins
You can implement every technical control and still get breached if your culture doesn’t support security. The final piece of AI security …
'I Only Built a Small Script for Myself.' That Might Be the Most Dangerous Sentence in Your Company.
·679 words·4 mins
35% of developers access AI coding tools through personal accounts. AI lets one person bypass every paved road the organization built, very fast and …
Your AI Browser Can Be Hijacked by a Single Webpage. Here's How Companies Are Fighting Back.
·2052 words·10 mins
AI browsers that summarize pages and automate tasks are vulnerable to prompt injection—hidden instructions in web content that can hijack the AI. …
Glassworm Is Back. Your Code Review Won't Catch It.
·924 words·5 mins
151 malicious packages in one week. The payload is invisible. Literally invisible. Glassworm uses Unicode characters that don’t render in any …
The Playbook
Hands-on: agents, RAG, MCP, and the tooling underneath — no hype.
AI Agents for Real Productivity: What Works in 2025
·2570 words·13 mins
Beyond the hype and the demos, what actually works when you build AI agents for real work? Here’s the landscape, the platforms worth using, and …
RAG for Developers: A No-BS Introduction
·1042 words·5 mins
Developers keep asking me to explain RAG. Here’s the straightforward explanation: it’s the difference between an AI that makes stuff up …
So, You've Built a RAG. Now Let's Make It Not Suck.
·2357 words·12 mins
Your basic RAG works in the demo. It falls apart with real users. Here’s how to upgrade from a fragile prototype to a production system that …
Model Context Protocol: The Missing Connection Between AI and Your Real Work
·3093 words·15 mins
Your AI coding assistant is blind to your company’s actual context. MCP fixes that. Here’s how to connect Claude, ChatGPT, and Cursor to …
The Magic Behind AI IDEs: How Cursor, Windsurf, and Friends Actually Work
·2145 words·11 mins
Everyone’s using AI IDEs but few understand what’s happening under the hood. Let’s demystify how these tools work, why they differ, …
The Context Problem: Why AI Can't Remember You Across Apps (And Why That's Not an Accident)
·1933 words·10 mins
Every time you switch from Claude to ChatGPT, you start from zero. It’s not a bug. It’s architecture. Here’s the real engineering …
Dispatches
News through an operator's lens — what launches mean for how orgs run.
I Don't Put All My Eggs in One Basket. Anthropic Is Making That Hard.
·2181 words·11 mins
Anthropic shipped 120+ features in 90 days, then blocked OpenClaw from using Claude subscriptions. The same company building the best developer tools …
Cursor Automations: Your AI Just Stopped Waiting for Permission
·712 words·4 mins
Cursor shipped Automations on March 5. AI agents now trigger from Slack messages, Git pushes, PagerDuty alerts, and timers. No human in the prompt …
Claude Can Now Use Your Computer. Here's What That Actually Means.
·826 words·4 mins
Anthropic just shipped computer use for Claude. It can click, scroll, navigate your browser, open files, run dev tools, and submit PRs. Pair it with …
GitHub's Double CLI Release: How Two AI Tools Are Reshaping Development Workflows
·2648 words·13 mins
GitHub released two different CLI tools for AI in one week. Together, they represent both interactive AI partnership and autonomous development …
The Agentic Commerce Protocol: We Just Gave Every LLM the Ability to Buy Things
OpenAI and Stripe just released a new protocol that lets AI agents complete purchases. This is the HTTP moment for AI commerce, and we need to talk …
The Uneven Reality of AI Adoption — What Anthropic's New Report Tells Us
·1187 words·6 mins
AI adoption has exploded from 20% to 40% of U.S. workers in just two years. But Anthropic’s new Economic Index reveals a troubling pattern: the …
By date 69 notes · newest first
202632
- Rethinking the SDLC: Execution Is No Longer the Constraint
- Everyone Trained Their Engineers on AI. The Gap Didn't Move.
- Make the Cheap Path the Default. Make the Expensive Path Prove It.
- Alberta Scanned 466 Million Lines of Code in 20 Hours. The Architecture Is the Story.
- Stop Reviewing Code. Start Reviewing Evidence.
- Agentic Overwatch: Why Your Next Dev Team Will Look Like a NASA Control Room
- The One-Man Show Company. Don't Let the Monkeys Touch Production.
- 100 Days to the EU AI Act Deadline. Your Engineering Team Hasn't Started.
- The Vibe Coding Backlash Is Right. Seniors Are Losing the Argument Anyway.
- The End of Courses: Learn From AI Like a Toddler, Or Become Obsolete
- I Don't Put All My Eggs in One Basket. Anthropic Is Making That Hard.
- The IDE Is Becoming Mission Control
- Your AI Stack Is Rented Until You Can Run Part of It Yourself
- AI Makes Code Cheap to Produce. Not Cheap to Own.
- 'I Only Built a Small Script for Myself.' That Might Be the Most Dangerous Sentence in Your Company.
- The Claude Code Leak Isn't Dramatic. That's the Point.
- Cisco Built an LLM Security Leaderboard. You Should Care Even If You Don't Use Cisco.
- OpenAI Killed Sora. That Tells You Everything About Where AI Is Actually Heading.
- Claude Can Now Use Your Computer. Here's What That Actually Means.
- DeerFlow 2.0: ByteDance Just Open-Sourced What Most Companies Are Trying to Build Internally
- Zuckerberg Is Building an AI CEO Assistant. The Rest of Us Should Have Started Already.
- Cursor Automations: Your AI Just Stopped Waiting for Permission
- AI Is Now Reviewing AI's Code. That Should Make You Think.
- Glassworm Is Back. Your Code Review Won't Catch It.
- WordPress Just Let AI Agents Publish to 43% of the Web. Now What?
- AI Didn't Replace Software Engineering. It Made Bad Engineering Easier to Ship.
- SaaS Is Dead. We Just Haven't Stopped Paying for It Yet.
- Google Stitch Just Made UI Design a Developer Skill
- OpenClaw Is Not a Chatbot. It's a Personal Agent Gateway.
- Open WebUI Isn't a ChatGPT Clone. It's AI Infrastructure.
- NotebookLM Is Not a Chatbot. It's a Research Workbench.
- Your AI Agents Are Flying Blind. Here's How to Fix That.
202537
- Org Charts for AI Agents: Mapping Your Human and AI Workforce
- Web Bot Auth: Giving Bots a Crypto ID Card in a World of Fakes
- CLI Agent Orchestrator: When One AI Agent Isn't Enough
- AI Wrapper Companies: Is This Real or Just API Theater?
- Amazon S3 Vectors: When Storage Learns to Think
- Your AI Browser Can Be Hijacked by a Single Webpage. Here's How Companies Are Fighting Back.
- Have You Seen All These OpenAI Blueprints? What the Heck Are They Doing, and Why Is (or Isn't) Your Country In?
- Krakow Offsite: Real Connections, Real Momentum
- When Nvidia's CEO Says 100% of Engineers Use Cursor, He's Not Exaggerating
- When AI Writes 90% of Your Code, What Are You Actually Doing?
- Securing Intelligence: The Complete AI Security Series [Video]
- AI Security Isn't a Tool Problem, It's a Culture Problem
- Securing the AI Supply Chain: The Threat Nobody's Talking About
- Building AI Systems That Don't Break Under Attack
- Prompt Injection 2.0: The New Frontier of AI Attacks
- AI's Dual Edge: When to Disrupt and When to Compound
- Grokipedia and the New Era: When Building a Wikipedia Becomes Trivially Easy
- Two Weeks with Gemini in Chrome: The Browser That Actually Gets It
- Google Jules: Always on My Radar, But Never Quite the Star
- Ship Faster Without Breaking Things: DORA 2025 in Real Life
- Build Your First AI Agent This Week: A Practical Guide
- AI Agents for Real Productivity: What Works in 2025
- What's Holding You Back from Succeeding in the AI Era?
- Model Context Protocol: The Missing Connection Between AI and Your Real Work
- The Context Problem: Why AI Can't Remember You Across Apps (And Why That's Not an Accident)
- The Agentic Commerce Protocol: We Just Gave Every LLM the Ability to Buy Things
- The Magic Behind AI IDEs: How Cursor, Windsurf, and Friends Actually Work
- Developer Work Did Not Change. The Sequence Did.
- GitHub's Double CLI Release: How Two AI Tools Are Reshaping Development Workflows
- So, You've Built a RAG. Now Let's Make It Not Suck.
- Hiring Developers in the Age of AI: What Actually Matters Now
- RAG for Developers: A No-BS Introduction
- The Uneven Reality of AI Adoption — What Anthropic's New Report Tells Us
- The Context Engine: What Comes After We've Solved Code Generation
- I'm Pro-AI. That's Exactly Why I'm Worried About Our Next Senior Engineers
- When CI/CD Speaks Human: A Friendly Nudge to DevOps (and Developers)
- From "Toys" to "Tools": The Missing Layer Developers Actually Need