fighting slop with slop — Vaibhav Gupta, Boundary
Vaibhav Gupta from Boundary presents a strategy for handling problems with AI agents that generate incorrect code and data—he calls it "fighting garbage with garbage." His approach uses cheap AI agents themselves as tools to monitor other agents: they review transcripts, flag hallucinations (when AI invents things), and compare which solutions worked best.
He combines this with stable rules and checklists that don't change, creating a solid foundation beneath the messy detection layer. The deeper strategy is to use type systems—design patterns that make certain errors impossible from the start—so agents can move quickly within defined boundaries without getting lost. He also presents BAML, a tool that lets agents work across Python, TypeScript, and Rust with strong constraints.
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