Evan Roubert

AI/ML Engineer focused on developing, training, and fine-tuning machine learning models for natural language and speech processing. My work includes: Large Language Models (LLMs) Automatic Speech Recognition (ASR) Audio Speech Synthesis (TTS) Speech-to-Speech (S2S) systems Dataset creation, preprocessing, and curation Model training, evaluation, and optimization Python development for AI research and production I primarily work with Python, PyTorch, Hugging Face Transformers, Accelerate, PEFT, DeepSpeed, and modern GPU training pipelines. I share datasets, models, and tools for the open-source AI community while exploring efficient training methods and real-world applications of language and speech models.