Rubrics as Reward: RL Beyond Verifiable Domains with Scale AI

​'Gradient Descending' is a series of roundtables on AI themes like MCP, Browser Agents, Evals and more.

​Rubrics as Reward: Reinforcement Learning Beyond Verifiable Domains. This will be a discussion on how RL can be applied beyond just objectively verifiable domains (like coding), with Aidan Davies (Solution Engineer at Scale AI) and Matthew Spaul (Strategist at Scale AI).

​​Sign up for future technical deep dive events and insights: https://www.akashbajwa.co/