Daniel Leonte

Projects

• SciQuery LLM: An advanced research assistant that answers complex AI research questions using Retrieval-Augmented Generation (RAG). Built with Sentence-BERT, FAISS, and Hugging Face’s DeepSeek model, it retrieves relevant arXiv papers and provides cited answers via a Gradio web interface. [Code]
• SpecialtyTuner: A lightweight, instruction-tuned language model for generating production-ready Python code for ML and data science tasks. Fine-tuned CodeLlama-7B with LoRA and quantized for efficient inference, achieving 95% syntax correctness. [Code]

About Me

I’m Daniel Leonte, a Senior Software Engineer with over five years of experience, mastering full-stack web development and advancing in machine learning. I’m pursuing B2B ventures to craft bold, meaningful projects with brilliant teams.

Contact / Links

• Email: daniel@danielleonte.com
• GitHub: daniel-leonte
• LinkedIn: Daniel Leonte
• Twitter / X: @danielleonteX