SimulationMaxxing: How we ship agents 20× faster — Aman Gupta (Nubank) + Shreya Rajpal (Snowglobe)
Nubank, which serves 135 million customers, used simulated training to deploy five AI agents to production 20 times faster than before.
The challenge was that testing agents took considerable time—they needed real data from customer conversations that were long and context-dependent, not just simple questions and answers. Snowglobe solved this by automatically generating simulated customer conversations with realistic details such as customer account information, tone, and purpose. This method worked well enough—simulated conversations matched real ones in approximately 80 percent of cases—to allow Nubank's team to test agents quickly and iterate without waiting for feedback from production.
Vibekollen prepared this summary with AI from the original publication. The content belongs to AI Engineer.
More from AI Engineer
Inside Arca's Long-Horizon Agents That Run the Business - Aman Singh, Arca
AI Engineer 23 Oct
Real AI Agent Attacks and What Would Have Stopped Them — Andrey Rybka, Bloomberg
AI Engineer 23 Oct
Outside of tech there has been no financial payoff to US companies for using AI, now viewed as a margin-hurting cost center, as VC-backed firms burn more tokens
Techmeme 4 h ago
Source: Replit recently moved most AI spend from Anthropic to OpenAI, which gave 30%+ off in a deal worth $100M+/year, after Anthropic discontinued its discount
Techmeme 6 h ago