A field guide to writing your own eval harness
Why "vibes-based" testing collapses past 50 prompts, and the smallest harness that scales without becoming a second product.
APR 28, 2026 ·
Building effective agents
"The taxonomy I keep coming back to. Workflows vs. agents, with worked patterns."
via Anthropic APR 26, 2026 ·
How we're shipping with autonomous agents at scale
"A pragmatic field report. The section on guardrail budgets is worth the read alone."
via Medium APR 25, 2026 ·
Benchmarking 7 coding agents on a real refactor
Same 12k-line TypeScript codebase, same task: extract a domain layer. I ran every agent twice and graded the diffs.
APR 22, 2026 ·
SWE-bench Verified — leaderboard notes
"Worth reading the verified subset methodology before quoting any number from the headline board."
via MarkTechPost APR 20, 2026 ·
The new contract between developers and AI
"Skim the intro, read the middle. Their framing of "negotiated autonomy" is sticky."
via The New Stack APR 18, 2026 ·
Things I have learned about LLMs in 2025
"Annual roundup. Section on tool-calling reliability is gold."
via simonwillison.net APR 16, 2026 ·
The shape of a good MCP server
Most MCP servers I see are CRUD wrappers. Here is what changes when you design tools as if a model were a junior engineer with no memory.
APR 12, 2026 ·
Subagents and the verifier pattern
"Forking a verifier you never await is the cheapest CI you will ever ship."
via Anthropic Docs APR 05, 2026 ·
Latency budgets for agentic UIs
A 90th-percentile breakdown of where the seconds actually go in a tool-using chat. Spoiler: it is not the model.
MAR 30, 2026 ·
Aider — repo-map heuristics, annotated
"The PageRank-on-symbols trick is more interesting than the headline feature."
via aider.chat MAR 24, 2026 ·
Designing for the model in the loop
When the user is not the only one reading your UI. A short manifesto on machine-legible interfaces.
MAR 19, 2026 ·