I'm a software engineering student and developer interested in backend architecture, fullstack development, AI/LLM integration, automation, and local AI character systems.

Currently, I'm building Maria — a local AI character engine / desktop companion project focused on personality, memory, emotional state, relationship dynamics, and long-term interaction.

• 🎓 Software engineering student
• 💻 Interested in backend, fullstack, AI systems and automation
• 🛠️ I like building projects that combine code, behavior logic and system architecture
• 🤖 Currently focused on local AI characters, LLM pipelines and character-based agents
• 📚 Improving my skills in OOP, testing, clean backend architecture and project structure

Maria is a local AI character project built around:

• persona and behavior rules
• memory and context systems
• affection / trust / mood mechanics
• relationship and emotional state dynamics
• opinion evolution
• local LLM integration through Ollama
• API layer for external clients
• future Unity / Live2D integration
• regression tests for behavior stability

The goal is not just to create a chatbot, but to build a more realistic character system with presence, personality and long-term interaction.

The screenshots and frontend showcase are available in a separate repository:

Maria AI Character Engine Debug WebUI

This WebUI is a developer/debug cockpit for Maria AI Character Engine. It is not the final user-facing Live2D/Unity client, but it demonstrates how the backend runtime can be observed and tested through a lightweight React/Vite interface.

It includes:

• live conversation view;
• daily dialogue history;
• runtime/debug state panel;
• relationship and mood parameter display;
• proactive message support;
• sprite overlay preview;
• API integration with the Maria backend.

This main repository focuses on the backend engine, architecture, tests and runtime logic.
For visual screenshots and the current web interface, see the Debug WebUI repository.

• Building Maria, a local AI character engine
• Improving backend architecture and OOP skills
• Learning testing, regression safety and cleaner project structure
• Exploring local LLMs, Ollama, Unity and Live2D integration

• GitHub: @HontoUKI