I hope you enjoy the fun. Everyone challenge was cool in its own right, but some were more fun.
This post is less humour, more plain tech speak. I’ll be back with more laughter in the next post. What…
...Three months later, older, wiser, frontier model powered.
Fun building LLM powered research agents.
Adaptive intelligence, and PII redaction.
Avoiding jailbreaks. No prisoners harmed in the writing of this.
The joys of testing, vibe coding problems and explaining the new architecture properly.
My AI now delegates confusion. Progress, apparently.
Who's the daddy? Decomposition and response answerers.
Vibe coding needs testing, and the importance of "Intent" resolution in retrieval.
Feeling the vibe, when fat gets thrown out for thin.