A Blog post by Build Small Hackathon on Hugging Face

HuggingFace's platform is a resource for developers and researchers working in natural language processing (NLP) and machine learning, offering insights into NLP models, tools, and datasets. Through articles, tutorials, and open-source projects, HuggingFace offers insights into state-of-the-art NLP techniques, transformer architectures, and transfer learning methods. Developers can learn about using pre-trained models, fine-tuning strategies, and deploying NLP applications with HuggingFace's libraries and APIs.

Hugging Face

A developer fine-tuned Google's Gemma 3 12B model using 16-bit LoRA via Unsloth to create NeuroBait, an ADHD-focused AI assistant that replaces overwhelming to-do lists with short, warm, dopamine-sparking nudges. Trained on a hand-curated synthetic dataset reflecting real ADHD friction, the fine-tuned model drops clinical structure and speaks in flowing prose, giving users one tiny actionable step. The model is deployed on Hugging Face Spaces using ZeroGPU with Gradio, bitsandbytes 4-bit quantization, and a LoRA adapter applied at runtime. Future plans include open weights, bilingual support (Indonesian and English), and community-driven development.

NeuroBait: I fine-tuned a model to spark dopamine for ADHD brain