Why it matters
It’s structured for people who already prompt LLMs daily but haven’t built anything that runs without them — the gap between ‘using a model’ and ‘operating an agent fleet’ that decides whether spend scales linearly or exponentially.
The tokenmaxxing angle
The framing explicitly names Hermes Agent as free/open-source and pairable with a local model via Ollama specifically to cut per-token API charges — a concrete build-vs-buy-inference decision rather than the usual 'switch to a cheaper API' advice.
From the organizers
The listing names Brad Freels as leading the guided lab and lays out a 'Crawl, Walk, Run, Sprint' build progression, ending with attendees assembling a small team of specialized agents; it's hosted by The AI Implementation Club at Antler VC's Austin office.