November 8, 2026 • 1 min read
Mapping the Largest LLMs: US vs China (November 2026)
A map of the largest language models from US and Chinese labs, ranked by estimated parameter count, in one table.
November 7, 2026 • 6 min read
Two Ways to Build an AI Research Workflow: NotebookLM + Claude vs. Claude Code + Jupyter
NotebookLM with Claude over MCP gives you managed document research. Claude Code with Jupyter gives you a programmable workspace. How they differ, where each fits, and when to combine them.
November 1, 2026 • 7 min read
Loop Engineering: Four Loops That Turn an Agent into a System
Prompt, context and harness engineering shape one run. Loop engineering covers how an agent acts, checks its work, reacts to events and improves over time.
October 29, 2026 • 5 min read
Treat Your Coding Agent Like a Naive Junior Developer: Four Layers to Protect Your Secrets
An AI coding agent is capable but has no security filter. Isolate the workspace, keep real keys out of reach, guard the pipeline, and scan every change.
October 27, 2026 • 6 min read
.env vs Vaults: Should an AI Agent Ever See Your API Keys?
Secret managers get keys off disk. Credential brokers keep them away from the agent. How Agent Vault, Aembit, AgentCore, HashiCorp Vault, Doppler and 1Password compare.
October 13, 2026 • 9 min read
Preflight Checklist to Build and Run AI Agents
A pilot-style preflight checklist for AI agents: capability leases, deny-by-default tools, evals first, attested MCPs and skills, in-flight guardrails and rubrics, cost routing, vendor-multi, reproducibility, and the rest of what must be true before takeoff.
October 12, 2026 • 11 min read
Deterministic First: Where AI Authority Ends
Where AI authority ends: a reference architecture for retrieval-backed, human-approved workflows, with three boundaries, a state contract, and a tested approval gate that rejects a recipient swap.
October 12, 2026 • 8 min read
Progressive Disclosure for AI Agents: Load the Skill, Not the Entire Playbook
How SKILL.md files let agentic applications, coding harnesses, and assistants like OpenClaw load specialized knowledge only when the task needs it.
October 3, 2026 • 30 min read
A Decision Model as a Reflex: A 10-Drone Simulation With and Without a Safety Backstop
Can a decision model acting as a reflex manage traffic and battery for 10 drones on a shared map? A 30-minute experiment with and without a deterministic safety backstop: 12 collisions without it, none with it.
October 3, 2026 • 8 min read
LLM Routing: Affinity, Caching, and Gateways
How generative AI routing differs from load balancing: KV caching, model affinity, switching budgets, and when to use an inference router, an AI gateway, or both.
Loading more articles…