"Forward to Shraga!" — Training a Small Agent to Handle 1M Production Logs a Day

In Ramzor, Hefer survives a high-tech job with zero technical skills using one move: forward everything to Shraga. It works beautifully until Shraga relocates to Silicon Valley and Hefer quits, because the whole system was Shraga. Most teams shipping AI features today are Hefer: every request gets forwarded to a big commercial model, and it works right up until latency, cost, privacy, or a deprecation notice takes Shraga away. This talk is about building your own Shraga — a small model fine-tuned into a capable agent for one specific job. We built an agent that reads IT logs, diagnoses failures, and proposes verified fixes. We’ll take it from a vanilla small model that fails almost on everything to one that holds its own. We'll walk through what it took to get there, while tackling two questions that should come before any training: do we even need an agent for every task? and if so — how do we actually evaluate one?