All Articles
Don't Trust the Confidence Score: Per-Class Calibration and Abstain Routing for Vision Classification
Overall accuracy looked fine while individual categories quietly collapsed. Here is how I calibrated Claude Vision's self-reported confidence per class and routed abstentions to a human queue.
Carrying Decisions Across Compaction with PreCompact and SessionEnd Hooks
Auto-compaction does not delete your conversation. It deletes the reasons behind it. Here is a working PreCompact / SessionEnd / SessionStart hook pipeline that rescues decisions to disk and hands them to the next session, with real code and measurements.
My Nightly Batch Was Quietly Running on a Bigger Model — Declaring Per-Task Model Ceilings and Enforcing Them
Scatter model names through your code and a cheap batch job will eventually run on an expensive model. Here is a manifest that declares a ceiling per task, a deny-by-default resolver, and a morning audit that catches drift — with working code.
Same Output, Different Path — Guarding Agent Trajectories with Invariants
When the default model changes, your final output can stay correct while the path your agent takes quietly shifts. Here is a trajectory regression harness built on recorded tool traces and deterministic invariants, with working code and measured numbers.
Is the Draft That Passed the Gate the Same One You Published?
In unattended pipelines, the file your quality gate inspected and the file you actually publish can quietly diverge. Here is a digest-bound gate receipt design, with working code and measured results from 180 days of running it.
When the RAG Started Being Confidently Wrong — Field Notes on Measuring Retrieval Misses With Groundedness
In a Claude API RAG, the answers stay fluent while the facts drift. Often the cause is a silent recall decay on the retrieval side, missing the document that holds the answer. Field notes on measuring groundedness and retrieval hit rate and walking the system back, with working code and real numbers.
The 200K-Token Cliff That Doubled My Nightly Bill — A count_tokens Preflight to Avoid Long-Context Pricing
Sonnet 5's native 1M context switches the entire request to long-context pricing the moment input crosses 200K tokens. Here's how I caught that silent cost cliff before sending, using the billing-exempt count_tokens API, with working Python and TypeScript code.
Claude Code vs Cursor: The Definitive 2026 Comparison — Choosing the Right AI Coding Tool
A comprehensive comparison of Claude Code and Cursor across pricing, features, accuracy, and workflow. Find the AI coding tool that best fits your development style with our 2026 data-driven guide.
Claude Credit Card Declined: How to Fix Payment and Billing Errors
Getting your credit card declined when subscribing to Claude Pro or Max? This guide covers every common cause—international transaction blocks, billing address mismatches, Stripe restrictions—and gives you clear steps to fix it.
Claude Card Declined: A Complete Troubleshooting Guide for Pro, Max, and API Users
When Claude tells you 'Your card was declined,' the cause is rarely obvious from the error text alone. This guide separates the three layers where a decline actually happens — your issuing bank, Stripe's fraud engine, and Anthropic's account state — and walks you through fixes plus fallback payment methods that almost always get the charge through.
Auditing Your Own Code for Required Reason API Declarations with Claude Code
Required reason codes in PrivacyInfo.xcprivacy apply to your own code, not just third-party SDKs. After a wallpaper app of mine was rejected, here is the Claude Code workflow I use to scan Swift sources and match APIs to reason codes before submission.
Rolling Out Agent Behavior Changes Gradually with Feature Flags
When an unattended agent changes its prompt or behavior all at once, quiet regressions hide until morning. Here is a design for assigning new behavior to only a fraction of runs, keeping a control group, and adding deterministic bucketing, canary comparison, and automatic rollback.