LLM Trainer and Finetuner
TextFine Tuning
I designed and trained a large language model (~0.5B parameters) from scratch, focusing on computational efficiency and resource optimization. The process involved preparing and curating datasets for training, finetuning the AI, and iteratively evaluating model performance. This experience required technical understanding of deep learning and careful management of labeled data for supervised learning. • Prepared datasets for AI training and evaluation
• Implemented quantization techniques to reduce memory usage
• Used Mixture of Experts architecture for scalability
• Compared inference performance with leading models
I designed and trained a large language model (~0.5B parameters) from scratch, focusing on computational efficiency and resource optimization. The process involved preparing and curating datasets for training, finetuning the AI, and iteratively evaluating model performance. This experience required technical understanding of deep learning and careful management of labeled data for supervised learning. • Prepared datasets for AI training and evaluation • Implemented quantization techniques to reduce memory usage • Used Mixture of Experts architecture for scalability • Compared inference performance with leading models
2024 - Present