Posted inBlog
What It Takes to Know If Your AI Agent Actually Works
For a while, the standard way to check whether an AI agent was working was to run it a few times, watch the outputs, and decide it looked reasonable. That…
Posted inBlog
Improving Agents Is a Data Mining Problem
Early agent development looked a lot like prompt alchemy: tweak the instructions, rerun the task, repeat. That worked at small scale, but as agents handle real production workloads and accumulate…
Posted inBlog
The Floor Is the Frontier: Why Agent Reliability Beats Benchmark Chasing
For the first couple of years of serious agent development, most of the energy went into the ceiling: what's the most impressive thing this agent can do? Can it beat…