Yu connects Vesta's growth with lenders adopting a new loan system and becoming more confident in AI-assisted work. He argues that mortgage origination combines complex knowledge work, established quality checksEvaluation measures how well an AI system performs against defined tasks, criteria and failure conditions using repeatable evidence. and a clear cost-per-loan measure, making the impact of automation easier to evaluate.
The agents take first passes at identifying documents and extracting fieldsIntelligent document processing uses AI to classify, extract and validate information from documents for downstream workflows., coordinate with third parties and use deterministic tools for calculations such as income. Yu emphasizes small, bounded tasks and existing human verificationHuman-AI collaboration combines human judgment and accountability with AI speed, generation, analysis, and tool use in a shared workflow. rather than allowing language models unlimited discretion.
The conversation also covers customers investing in Vesta and the importance of long-term alignment. Yu describes a prospective personal-assistant layer that would help employees manage many task-level agentsA multi-agent system contains multiple AI agents that interact, coordinate, divide work, or influence one another while pursuing tasks. while preserving their role in loan oversight and communication with consumers.
Watch on YouTube
