Software Engineer 1
Current- Led the development of a multiclass text classification project utilizing a hybrid LLM approach.
- Managed end-to-end project lifecycle, including data preprocessing, text augmentation, and handling data imbalance through advanced techniques (under/oversampling, weighted sampling).
- Applied multiple feature extraction methods (TF-IDF, Word2Vec, BERT) to improve model inputs.
- Integrated Azure AI Search Index with a fine-tuned LLM, boosting classification accuracy to 90.5%.
- Spearheaded and successfully led initiatives in RAG (Retrieval Augmented Generation), contributing to advancement of content generation techniques.
- Applied state-of-the-art LLMs like OpenAI’s ChatGPT and Meta’s Code-Llama, Llama 2 for various applications.