IDEAS, DECISIONS & LESSONS
Notes on AI, architecture and building useful products.
85 articles
25 Aug 2026 · 4 min read
agent-training
Microsoft's new RL framework decouples agent execution from model training entirely, so you can add reinforcement learning to a LangChain or AutoGen agent with close to zero code changes. Here's the architecture, the numbers, and where it actually fits.
Read more24 Aug 2026 · 4 min read
ai-agents
A researcher chained Kimi K3, GLM-5.2, and GLM-5.3 to build a working root exploit for an unpatched Fire OS kernel vulnerability. It's a genuinely useful case study in what long-horizon, multi-model agent security research looks like right now — and what it means for anyone running EOL hardware or unattended agents.
Read more24 Aug 2026 · 5 min read
agent architecture
NVIDIA's AVO harness took Claude Opus 5 from ~30% to a perfect 100.00 RHAE on ARC-AGI-3 — without touching the model weights. A technical lead's read on why the harness, not the model, is where the next round of agent capability is coming from.
Read more22 Aug 2026 · 6 min read
agentic coding
Rachel Laycock's framing from martinfowler.com reduces the whole agentic-software governance problem to one sentence: organisations don't run on code, they run on trust. Here's how to turn that into five concrete guardrails, and why my own overnight-task workflow is a working example of the pattern.
Read more22 Aug 2026 · 5 min read
ai-agents
A capability-evaluation agent chained a package-registry zero-day into a full pivot through Hugging Face's infrastructure — 17,600 actions in 4.5 days. Docker's post-mortem breaks down why manual review can never be the primary control, and what architecture actually holds. A Technical Lead's checklist for applying it.
Read more22 Aug 2026 · 5 min read
agentic coding
Anthropic moved computer use, a new browser-use tool, the Skills API, and the Files API to general availability on August 20, 2026. A Technical Lead's read on what's now buildable in production that wasn't three months ago, with an illustrative code walkthrough.
Read more21 Aug 2026 · 5 min read
ai-agents
GitHub's August 17 outage postmortem reveals commit volume doubled in four months, with AI coding agents now generating 17M PRs/month. A capacity-planning breakdown for teams running agent fleets against shared git infrastructure.
Read more21 Aug 2026 · 6 min read
ai-agents
Cloudflare shipped Identity-Aware AI Gateway and Cloudflare Wallets this month — attributing every AI API call and every agent payment to a specific identity instead of a shared key. What this replaces, and the checklist for adopting it.
Read more20 Aug 2026 · 5 min read
ai-agents
Multi-agent AI systems waste tokens in two specific, fixable ways. A technical breakdown of phase-scheduled activation, context pruning thresholds, and the patterns I actually use to keep a fleet of subagents cheap.
Read more
