Agents Building Agents - Alfonso Graziano, Nearform
NearForm uses AI agents and spec-driven development to iteratively build more reliable, secure AI agents.
“an agent is an LLM inside a an agentic loop, which is connected to tools and can retrieve context. That's it.”
A NearForm tech lead presented a repeatable process for using AI agents to build AI agents, addressing failure classes like non-determinism, hallucinations, and poor eval performance. It is a practical engineering talk on agentic development workflows rather than a major industry announcement, offering useful but incremental insight.