03 / Writing
Blog
Stop Preparing for AI Job Loss Start Preparing for AI Meaning Loss The Real Crisis Isn't Unemployment—It's What Happens When Work No Longer Defines Us
Why the AI meaning crisis poses a deeper threat than technological unemployment, and how to prepare for a future where work no longer defines human purpose.
Your Code Knowledge Graph Needs Four Layers, Not One
Memory should be a graph. The graph should be four layers. The middle one carries navigation. The top one turns navigation into action.
A Sablier Bug That Wasn't Sablier: 4 Gotchas from Tracing a Traefik Plugin Failure
Tracing a sporadic 'invalid middleware' error in Sablier to a hidden startup dependency introduced by a Traefik 3.5.3 refactor.
Stop Putting Decisions in CLAUDE.md. Put Them Where the Agent Won't Always Read Them.
Two 2026 studies disagree about whether AGENTS.md helps coding agents. The fight makes sense once you separate push context from pull context.
Streaming UI from AI Agents: 5 Tools Ranked
AG-UI, A2UI, Vercel AI SDK, TanStack AI, and Kombai take different bets on how agents should build interfaces. The architectural question decides which wins.
Capital-of Is Not a Single SAE Feature. So I Built a Mutation Loop to Find What Is.
SAE features can't isolate relations in Gemma-2-2B. I built a mutation-selection loop that can. The bottleneck was tokenization.
The Bridge Is the Problem. Direct Access to the Divine Needs No Intermediaries.
Yoga, Gnosticism, Zen, Christian mysticism. They all show the same thing: direct connection was always there. Who blocked it, and how do you get back?
Your MCP Setup Is Burning 90% of Its Context Window. Here's the Fix.
Every MCP tool you connect loads its full schema upfront, before you type a word. Anthropic's deferred loading fixes this.
Sparse Autoencoders Can't Measure Generation-Time Behavior. That's Not a Bug.
Why sycophancy SAE features have Cohen's d=9.9 but hallucination detection fails. The answer turned out to be deeper than measurement timing.