Posts
All the articles I've posted.
The Decision Plane: Why Jev Changes Enterprise Agent Architecture
Published: at 03:00 PMMost enterprise agent traffic is decisions, not prose — yet every branch is produced by an autoregressive model. Here is what changes when the decision layer comes out of the LLM and back into code, and the cascade architecture the accuracy numbers actually justify.
Jev: The Frontier Model That Refuses to Write
Published: at 10:00 AMJev is a frontier model that never generates text. It scores the options you supply in a single parallel pass — 8.5 seconds of LLM output replaced by 0.1 seconds of arithmetic. Here's the trade, and why it matters for the cost architecture of every agent you run.
Cloud Agents Will Become the De Facto Way We Build Software
Updated: at 09:59 AMCloud agents like Claude's managed agents and Devin are moving AI coding from your laptop into hosted environments that run 24/7. Here is why cloud-hosted agents will become the default way software gets built—and why verifying their work is the skill that separates teams that ship from teams that break.
AI Software Factories - The Industrialisation of Software
Updated: at 09:59 AMWe are entering the industrial age of software engineering. Agent factories in the cloud will write 60-90% of code, fundamentally reshaping the SDLC and the role of human engineers.
Dense vs MoE - Choosing Between Dense Models and Mixture of Experts
Updated: at 10:59 AMDense vs MoE explained: how dense and Mixture of Experts models differ in compute, memory, latency and cost, and a practical rule for when to run each, including on local hardware.
The CAIO's First 90 Days: AI Governance Frameworks That Actually Stick
Published: at 11:00 AMA practical guide for newly appointed Chief AI Officers on the essential AI governance frameworks to establish within the first 90 days, covering NIST AI RMF, ISO/IEC 42001, and minimum viable governance structures.
How to Present an AI Strategy to Your Board (Without Losing Them in 5 Minutes)
Published: at 11:00 AMThe bridge between your deep technical knowledge and boardroom clarity. Frameworks, visuals, and language that resonates with non-technical directors when presenting AI strategy.
Agentic Engineering Patterns- The Foundation of AI-Native Development
Updated: at 10:59 AMDiscover the three pillars of agentic engineering patterns: agent harness, content engineering, and compound engineering—the blueprint for building AI-native systems.
C-Suite Leaders Must Embrace Rapid Experimentation in the AI Era
Updated: at 12:59 AMIn the AI era, traditional strategic planning cycles are obsolete. C-suite leaders must shift from exhaustive planning to rapid experimentation, embracing action and agency over endless analysis. Here's why experience is becoming a tax, not an asset.
Compound Engineering v3 and the Rise of Agentic Software Delivery
Published: at 09:00 AMHow Compound Engineering v3's improvements—unified namespaces, traceable requirements, and better tooling—enable the next wave of agentic AI and faster software delivery.