AI Engineering Is Context Management the New Memory Management? From malloc to meaning: how decades-old instincts about memory management find new life in the age of context windows and retrieval graphs.
Vibing to Production Vibing to Prod: Charting the Journey from Sparks to Systems Most innovation dies in a branch called experiment/. We want ours to ship. Vibing to Prod is how we plan to get there. It's a journey from sandbox to signal, guided by FLUID loops. Turning GenAI speed into sustainable delivery without losing the magic that started it.
Better Bets Better Bets, Not Bigger Burdens: Why "Agile+AI" Isn’t Delivering the Gains Most teams are “doing agile” but really just running faster on old rails. Layering AI on top of that doesn’t deliver transformation — it just adds noise. Real gains come from rethinking org design and process flow so AI fuels iteration, not bigger burdens.
Better Bets Better Bets, Not Bigger Burdens: Shaping Delivery with Specs, Prototypes, and Agent Factories Exploring how specs, prototypes, and agent factories are reshaping delivery in the AI era. Specs shape work, prototypes teach through fast feedback, and agent factories collapse cycle time — together enabling better bets without bigger burdens.
Better Bets Featured Better Bets, Not Bigger Burdens: How AI Redefines Delivery at the Edge AI is not just about faster delivery; it is about making smarter bets. By blending innovation with execution, companies can turn ambiguity into clarity, reduce risks, and reshape product direction without adding bigger burdens to teams.
Better Bets Featured Better Bets, Not Bigger Burdens: Rethinking AI-Accelerated Delivery AI-assisted development isn’t about doing more. It’s about learning faster, failing safer, and placing smarter bets. Don’t burn your gains, instead invest them in discovery.
AI Augmentation Featured AI and the Death of the Hero: A Jungian Guide to Letting Go of Control As AI continues to reshape how we build and think, resistance remains strangely personal. Why do some people embrace the change while others cling to old ways? The answer may lie deeper than tooling or training it may live in our collective psyche.
Vibe Coding AI-Assisted Development: Beyond Code Generation AI-assisted development isn’t just about code generation. From debugging and scaffolding to test creation and doc writing, AI accelerates how we explore, explain, and execute ideas. Here are seven practical ways AI is quietly reshaping real developer workflows.
LLM as Judge How LLMs Judge: 4 Essential Patterns for Smarter Agent Workflows The LLM-as-a-Judge pattern originated in eval pipelines to assess output quality, but has since expanded into agent architectures. It now plays a central role in autonomous workflows critiquing outputs, enforcing standards, and triggering rework steps to ensure runtime quality control.
Vibe Coding Learn Fast. Think Independently. F*** the Rules. AI is reshaping software development. This piece challenges us to embrace the shift—away from rigid processes and toward fast, intent-driven workflows. It's a call to rethink how we build, lead, and ship in an era where AI isn’t just a tool, but part of the team.
Vibe Coding Featured AI Is Step One. What Comes Next Changes Everything. AI-assisted development isn’t the destination—it’s step one. Software is one of the most labor-intensive human creations, and AI is beginning to reduce that burden. But the real shift lies ahead: adaptive systems, dynamic governance, and automation that builds automation.
Vibe Coding Featured AI Didn’t Kill Coding. It Killed Typing! Typing was never the work—solving problems was. AI pair-programming skips boilerplate, letting devs focus on architecture, edge cases, and impact. Welcome to vibe coding, home of the 10× engineer.
LLMs Featured Are We Just Language Models in Meatspace? Maybe you’re not so different from an LLM. You just hallucinate with more confidence. This piece unpacks the “coherence trap,” why AI feels smarter than it is—and why that might reveal something uncomfortable about us.
Workforce AI and the Vanishing Ladder: Rethinking Work in the Age of Automation AI is erasing entry-level roles, gutting the talent ladder. Juniors aren’t learning, orgs lose future leaders, and inequality grows. To thrive, leaders must protect onramps, embrace hybrid roles, and invest in human development—not just automation.
Coherence Featured The Coherence Trap: Why LLMs Feel Smart (But Aren't Thinking) LLMs don’t think—they resonate. In this article, I unpack The Coherence Trap' the illusion of intelligence created by structured, fluent language. We explore how meaning emerges from alignment, not understanding, and why designing for coherence—not cognition—is the key to building with AI.
AI Agents Agentic Protocols: Cutting Through the Noise Agentic protocols like MCP, A2A, and ACP aim to standardize how AI agents interact—but each serves different needs. MCP excels at structured execution, A2A at distributed coordination, and ACP at simple integrations. Choose based on context; they’re tools, not truths.
HumanAugmentation Featured Prompting Is Dead. Long Live the Loop. Prompting alone won’t cut it. By reframing AI work as a structured loop of context, feedback, and iteration true collaboration emerges not from clever one-liners but from designing fluid, adaptive interactions where meaning is shaped together.
Vibe Coding Featured Vibe Coding: The Automation Moment for Software Vibe coding is the automation movement for software. It shifts the focus from typing to thinking, collapsing ceremony and exposing what really matters—design, validation, and impact. Coding isn’t dying, it’s evolving.
Coherence Featured Coherence Confirmed: LLMs Aren't Thinking, They're Coherent LLMs aren’t reasoning—they’re aligning. A new ICLR paper backs my theory: coherence, not cognition, drives meaning. These models flow through context, not thought. That shift changes how we prompt, evaluate, and design with AI.
Utility Featured Coherence, Not Cognition: Rethinking How LLMs Help Us Think LLMs don’t recall facts—they reconstruct meaning. This blog introduces “coherence reconstruction,” a practical theory explaining why LLMs feel useful even when wrong—by simulating thought-like patterns that amplify human reasoning and creativity.
Architecture Architect the Right Problem: Why Thinking Is the New Engineering In the AI era, code is no longer the bottleneck—problem framing is. This blog explores why the highest-leverage skill in modern engineering is architecting the right problem, not just solving one efficiently. Thinking is the new engineering.
AISoftwareDevelopment Mutable by Design: The FLUID Software Philosophy SOLID taught us structure, but AI-native development demands adaptability. FLUID is a new mindset for coding with machines — fast, collaborative, and mutation-friendly. This piece explores how to move from code as artifact to code as interface.
Vibe Coding Featured The Vibe Shift in Software: Why AI-Assisted Coding Is Here to Stay AI-assisted dev isn’t just autocomplete. It’s a shift from syntax to intent, from isolated tasks to conversation. Whether guiding agents or prompting apps, the real story is how we think and build with machines.
Decision Research Featured How I Unlocked Smarter GenAI Use: Lessons from 2,500 ChatGPT Conversations Analyzing 2,500 ChatGPT chats showed that AI’s true value emerges through engagement. Passive use leads to automation bias, but iteration and refinement unlock value and creativity. AI isn't an autopilot. It is a thought partner! The more we engage, the better our results.
AI UX The Touchless Car Wash Approach to GenAI: Rethinking UX for AI AI should be as effortless as a touchless car wash—seamless, adaptive, and requiring no manual fine-tuning. Most AI tools still demand large amounts of effort. This post explores how GenAI must evolve to provide true utility, removing friction and integrating smoothly into everyday workflows.
