Sam Charrington interviews Dan Jeffries, creator of Nautilo, about building personal and collaborative agents. Jeffries describes agents that keep separate contextsContext engineering designs how relevant instructions, evidence, state and tool information are assembled for a model's task., retrieve memoriesAI agent memory is stored information that an agent can retrieve and use across steps, sessions, or changing contexts. and participate alongside people in shared conversations.
They discuss separating probabilistic decisions from deterministic toolsAI tool calling lets a model request a structured operation from software instead of only returning natural-language text., the trade-offs between local and cloud modelsModel selection chooses the AI model whose capability, quality, cost, speed, safety, and operating constraints best fit a task., and protecting sensitive informationData privacy governs how personal, confidential, or sensitive information is collected, used, shared, retained, and protected in AI systems.. Jeffries also describes the difficulty of differentiating and marketing an agent product amid widely available general-purpose systems. His claims about his own product and predictions are presented as his perspective.
Watch on YouTube
