Sebastian Proactive: a local‑first AI companion that initiates conversations (github.com)

Sebastian Proactive, an innovative local-first AI companion, has been announced, utilizing local large language models (LLMs) through Ollama to initiate conversations with users. This proactive AI checks in based on user-defined schedules and dynamically understands natural time expressions. Key features include a three-tier memory system for managing fresh, medium, and long-term interactions, enabling Sebastian to evolve with users over time. The companion also implements a terminal-based interface, responds with various personality cues, and adapts its tone according to the time of day, enhancing engagement. This development is significant for the AI/ML community as it showcases advancements in local LLM applications, moving towards more personalized and context-aware AI interactions. The incorporation of auto-scheduling, a cue system for character personality variation, and robust error handling highlights the commitment to creating a responsive and tailored user experience. The system's ability to gracefully manage connection errors with Ollama further emphasizes the focus on reliability in AI interactions, making Sebastian a noteworthy step forward in the evolution of personal AI companions.