IDEAS, DECISIONS & LESSONS
Notes on AI, architecture and building useful products.
917 articles
26 Aug 2026 · 4 min read
ai-agents
Ollama v0.33.0 shipped a UI-driven gateway that lets Claude Desktop run Qwen, DeepSeek, and Kimi models locally — no MCP server, no Anthropic feature, entirely Ollama's own proxy. Here's how it works, the tradeoffs in tokens/sec and tool-calling reliability, and when it's worth flipping on for a team.
Read more26 Aug 2026 · 4 min read
ai-agents
Docker shipped microVM-isolated sandboxes that plug directly into GitHub Agentic Workflows, giving AI coding agents a full private Docker daemon and sudo without touching your runner's secrets or host. Here's the architecture, a working YAML example, and where the rough edges still are.
Read more26 Aug 2026 · 4 min read
agentic engineering
A one-line prompt tweak can silently break an agent, and you won't notice until a user does. My fix is an eval harness: a golden set of inputs and checks, run against every agent and workflow on every change. Here's the anatomy — deterministic checks, LLM-as-judge, run-on-change regression, and grading multi-agent workflows.
Read more26 Aug 2026 · 3 min read
architecture
Daily English practice for tech professionals. Morning session covering software architecture and scalability with pronunciation guide, exercises, and real-world examples.
Read more26 Aug 2026 · 4 min read
agent tools
A 'bot' in an agent platform isn't a monolith. It's a small tool plus a persona that knows when to call it. Using my Grok image generator as the worked example: the tool-vs-MCP decision, the function schema that doubles as a prompt, the tool loop, and turning a tool into a bot.
Read more25 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 more25 Aug 2026 · 5 min read
ai
Anthropic logged its 166th outage of 2026 this week, the sixth in August alone. If your product has an LLM call on the critical path, here's the resilience architecture that actually survives that.
Read more25 Aug 2026 · 5 min read
ai
GLM-5.3 just jumped from 4.6 to 28.3 on Terminal-Bench 3.0 at a fraction of frontier pricing. Here's the routing architecture we actually shipped to stop overpaying for tasks that don't need it.
Read more25 Aug 2026 · 4 min read
ai-ml
Daily English practice for tech professionals. Morning session covering prompting, retrieval-augmented generation, and grounding vocabulary — with pronunciation guide, exercises, and real-world examples.
Read more
