The Archive
Deep dives into AI agents, coding tools, and building with LLMs.
1209 articles844 topics
Goldman Sachs limited Claude Code and GitHub to engineers in October. What Bloomberg reported, what it signals for banks and regulated teams, and a rollout checklist for your own org.
llama-modchip claims DeepSeek-V4-Flash 284B fits in 64GB of RAM with no GPU. What the repo says, vendor-reported speeds, what Hacker News could not reproduce, and when a hosted model wins.
Apiiro found real secrets in 1,122 public Claude Code settings.local.json files saved by Yes, and don't ask again. What the file stores, what Anthropic said, and a 10-minute cleanup.
Claude Code 2.1.319 adds permissions.denyInsteadOfPrompting: any call that would wait on a permission prompt is denied and the turn carries on, subagents included.
Claude Console now has an IP allowlist in beta: up to 1,000 IPv4 or IPv6 ranges per org, enforced on the Messages, Batches, Files and Admin APIs. Setup and lockout traps.
Conductor is joining Meta. Cloud goes read-only Nov 13 and shuts down Nov 20, and the local Mac app winds down in early 2027. The dates, what to export, and what to move to.
Ricursive released Ripple-65B Preview, an Apache 2.0 dense model claiming US open-weight SOTA on Terminal-Bench 2.1. What the model card actually says, why HN pushed back, and who should test it.
Weekly Claude Code limits are draining faster for many users since mid-October. What the GitHub measurements show, the headless claude -p metering gap, and what to check first.
Developers report Codex model training turning itself back on. What OpenAI documents, the second "Include environments" switch most people miss, and a five-minute check.
A Tell HN says paying for Gemini does not stop chat training. What Google documents about Keep Activity, human review and 3-year retention, and what developers should do before pasting code.
Claude Code on Opus 5.5 now refuses to recreate specific existing games, even renamed. What the two-sentence rule says, what it blocks, and what builders do next.
Anthropic renamed Max 20x to Ultra and dropped the five-hour limit for verified accounts. What the 10x weekly label means, how to check yours, and Max vs Ultra.
Showing 12 of 1209 articles
Reading Paths
Start with the strongest evergreen pieces, then branch into the weekly notes, guides, and comparisons around the same problem.
Claude Code, Cursor, Codex, and the operating habits that make agentic coding useful in real repositories.
How to choose between orchestration frameworks, UI copilots, stateful agents, and production app patterns.
The practical layer for giving agents access to tools, files, services, and repeatable developer workflows.
Permissions, prompt injection, tool access, audit logs, and rollback patterns for agents that can touch real systems.
Model routing, AI SDKs, deploy platforms, and the infrastructure decisions behind reliable AI products.
How the site, app directory, public APIs, and content loops are built, wired, and improved in public.
Start Here
Series
Deep dives on AI agents, coding tools, and building with LLMs - delivered weekly. No spam.
Free forever. Unsubscribe anytime.
New tutorials, open-source projects, and deep dives on coding agents - delivered weekly.
