Technically Speaking

Write

The Claudish Tax: What Four Months of Opus Prose Cost Us, and How to Sweep It Out

Sep 30, 202612 min read

No. You Don't Need Cryptographic Hashing.

Sep 27, 202619 min read

The Clockwork Candidate: What Happens When Inbound Costs Nothing

Three applications, seventy minutes apart, identical echoes. Notes on the industrialization of synthetic hiring.

Oct 2, 20266 min read

How to Corral Gemini for Agentic Coding

Sep 25, 20264 min read

Forcing Claude Code to Speak Human

Sep 24, 20265 min read

Part 4: Real-Time Pipeline Diagnostics

Calibrated Flake vs. Regression Triage in Automated Builds

Sep 23, 20265 min read

Part 3: Governing the Inner Loop

Fast Probabilistic Guardrails for Autonomous Agents

Sep 22, 20266 min read

Building on Rented Intelligence: The Receipts Behind Claude's Silent Downgrade

For weeks, I've been watching Claude suddenly choke on fully specified engineering tasks it used to crush. I called it exactly what it felt like: Anthropic was silently lobotomizing their models. When

Sep 21, 20263 min read

Part 2: High-Speed Boundary Classification

Protecting Ingress with Calibrated Probabilities

Sep 21, 20265 min read

Part 1: The Missing Primitive in Agentic AI — Kicking the Tires on TypeSafe's Jev

The Missing Primitive in Agentic AI — Kicking the Tires on TypeSafe's Jev

Sep 18, 20267 min read

Resolving a Council of Independent AI Reviewers into a Single Decision

Running multiple frontier LLMs in parallel to evaluate prompt pipelines, codebases, or architectures is straightforward. The hard part is reconciling their outputs. When three independent agents—such

Sep 18, 20267 min read

Command Palette

Latest articles