The Developer's Role in the Age of AI
AI has cut the time it takes to write code, but responsibility for architecture, security, and correctness still sits with the engineer. How the developer role moves up the stack, and how to use AI without losing engineering ability.
Recent posts
View all →-
Agent Delegation Through the Lens of RIP and BGP
Agent handoffs beyond a fixed team are a distributed-state problem. Recorded paths, expiring claims, explicit withdrawals, and monotonic versions from routing protocols apply directly to delegation between agents.
-
Instrumentation Inherits Its Author's Blindspots
Self-instrumented systems are blind in a direction you can predict from who wrote the logging. Why standard telemetry misses attacker behavior, and how to log the application's assumptions and threat model to get the second reading.
-
From Threshold Alerts to Information-Theoretic Stability: Quantifying Behavioral Drift
Alert volume is the wrong thing to measure. Using information-theoretic divergence (Jensen-Shannon, KL) to quantify behavioral drift, and turn vulnerability management into continuous behavioral integrity.
-
From the Louvre to AFCON: A Structural Pattern
The Louvre ticket-fraud ring and AFCON 2025 scalping look unrelated, but they share a structure: when access isn't bound to verified identity, it becomes currency. A look at the wider MEA fraud economy.
-
The Day I Could Not Buy Lunch
A first-person account of the April 2025 Iberian Peninsula blackout, and what ten hours without power or connectivity reveal about how completely modern life now depends on systems we don't control.
-
Identity Isn't a Component of Security. It Is Security.
Identity has become the most fragile, least-governed layer of the enterprise. Why privilege sprawl and fragmented trust, not insufficient technology, drive modern breaches, and what actually reduces the risk.