video1.0<iframe src="https://www.loom.com/embed/c616603dcb8e40519cd175991bae6b4f" frameborder="0" width="1658" height="1243" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>12431658Loomhttps://www.loom.com12431658https://cdn.loom.com/sessions/thumbnails/c616603dcb8e40519cd175991bae6b4f-0621b64464fd6505.gif237.46133300000002This Loom describes a decision support project that uses real prior human outcomes to help with hard personal choices. It explains that online advising is opinion and the goal is outcome by finding precedent stories, turning them into a council of agents per group, and debating claims tied to their real sources without inventing gaps. The example decision is whether to move to a new city for a better job or stay close to family, which the model frames as career versus love and returns a verdict, selecting to move based on prior insights. The main challenges highlighted are scraping, using a tool like Epify with multiple scraping tools, and difficulty retrieving relevant results via keyword searches (for terms like stable job), while relying on RAG over fine-tuned models.