Topic
6 posts.
Interactively, Qwen Code never gave me trouble. The first time I ran it as a one-shot -p job, the answer was slow and a read-only df was refused. I followed it through the source to two bugs, filed both, and wrote a fix for the one that comes down to a single regular expression.
In my last Kong post I broke requests one at a time and read the evidence by hand. This time I asked what the gateway can tell me over hours and across several agents at once, and put its metrics next to the model server and the GPUs.
I put Kong Gateway and its AI Proxy plugin in front of my local LLM, measured what the extra hop costs, then broke requests on purpose to see where the evidence of each failure ends up.
A talk from IBM, Confluent and Google Cloud argued that AI agents fail for lack of fresh, refined context, not for lack of a better model. I rebuilt the idea with Kafka, Flink, MCP and a local LLM, then asked one question four ways.
Just for fun: a native 3D Rubik's Cube in Go and OpenGL that scrambles and solves itself. A local Qwen 3.8 Flash Next wrote all of it, with Claude Code as the harness.
A working demo of an event-driven AI system. Kafka carries what is happening, Flink works out what matters, and a local LLM investigates each alert and decides what the bank does next.