The Types Landed, but Nothing Got Safer — Where I Delegated a JS→TS Migration to Claude Code, and Where I Didn't
A green tsc run does not mean any is gone. Measuring a migration by type coverage, drawing a clear line between what Claude Code handles well and what a human must decide, and a CI ratchet that refuses regressions.
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.
Contract-Test Every Tool Before You Submit or Automate an MCP Connector
A connector that works once in a chat can still break silently in an unattended job through misread response shapes or double-fired writes. Here is a small harness that machine-checks tool descriptions, response contracts, idempotency, and latency, with measured numbers.
Five Minutes of Silence, and Something Retries on Your Behalf — Rethinking Retry Ownership After the Streaming Idle Watchdog Became a Default
Claude Code's streaming idle watchdog is now on by default, quietly adding another retrying layer to your stack. This article inventories the four layers (SDK, wrapper, watchdog, scheduler), computes worst-case attempt amplification, and shows how to collapse retry ownership into a single layer.
A 40% Lower Price Doesn't Mean a 40% Lower Bill — Measuring the Opus 4.8 to Sonnet 5 Migration by Cost per Completed Task
Sonnet 5's intro pricing looks ~40% cheaper than Opus 4.8, yet extra tool turns can flip the math. Working TypeScript for consumption vectors, a paired-run harness, and break-even turn counts.
When the Model Survives but One Parameter Expires: A Dated Deprecation Calendar for Claude API Requests
Your model ID can stay valid while a parameter you pinned quietly reaches its sunset date and takes the batch down with it. Here is a design that breaks a request into parts, gives each part its own expiry date, and catches the problem before the call goes out — with working TypeScript and real operational numbers.
When Claude API Document Extraction Is Confidently Wrong — Field Notes on Catching Silent Errors with Invariants
In structured extraction from invoices and contracts, the real danger isn't a crash — it's a value that's silently wrong while the schema validates and confidence reads high. Field notes on invariants, two-pass extraction, and tracking field-level error rates.
A Fail-Closed Model Pricing Registry So New Models Don't Quietly Break Your Cost Math
When Opus 4.8 and Haiku 4.5 landed in the Messages API, rates scattered across my code silently skewed the cost rollup. Here is how to centralize per-model rates and fail closed on unknown models, with complete working code.
When Your Claude × Playwright Browser Agent Fails While Reporting Success — Verifying Actions and Catching UI Drift
A vision-driven Claude + Playwright browser agent fails quietly: it reports success while nothing actually changed. Here is how to stop trusting self-reports, verify each action against the goal, and detect UI drift before it breaks you.
When Claude API Extracts the Wrong Value With Full Confidence — Designing the Verification Layer
When you extract invoices or contracts with Claude API, the scariest failure isn't an exception — it's plausible-but-wrong JSON. Here is how I build a verification layer that catches silent extraction errors with schema checks, arithmetic reconciliation, and dual-extraction agreement, in TypeScript.
Making the Numbers Add Up in a Multi-Tenant Claude API SaaS — Field Notes on Isolation and Cost Attribution
The first thing that breaks when you make a Claude API SaaS multi-tenant is the month-end reconciliation. Here are field notes on a single metering chokepoint, atomic counters, reconciling against Anthropic's bill, and proving tenant isolation with adversarial tests — with production TypeScript.
Beyond Tools in MCP: Designing with Resources, Prompts, and Sampling
Cramming everything into MCP tools hits a wall fast. Here is how resources, prompts, and sampling untangle a server, told through a real wallpaper-app asset manager I cut from 14 tools down to 5.