Experience

• [Apr 2023 - Present] Applied Scientist, Machine Learning @ Ownwell [Austin, TX]

  • Developed models for housing price prediction, document processing, and question-answering
  • Conducted prompt engineering and fine-tuning of Large Language Models (LLMs) to optimize performance and adapt them for custom tasks
  • Collaborated with cross-functional teams to brainstorm and successfully integrate new AI functionalities into the product
  • Utilized machine learning techniques to automate and significantly improve the efficiency of internal and external processes
  • ML Stack: Transformers, multi-modal fusion, gradient boosting, and temporal self-attention
  • Tech Stack: Python, PyTorch, Scikit-learn

• [Jun 2022 - Sep 2022] Applied Scientist Intern @ Amazon [Seattle, WA]

  • Developed ML models to detect fraud from user behavior patterns (mouse, keyboard gestures) and browsing data for the Buyer Risk Prevention (BRP) Team. Improved the performance of the production model by 6.96%
  • ML Stack: Temporal models such as LSTMs, multi-modal fusion, gradient boosting, and temporal self-attention
  • Tech Stack: Python, PyTorch, Scikit-learn

• [Jun 2021 - Aug 2021] Research Intern @ Nokia Bell Labs [Remote]

  • Developed a self-supervised framework for extracting features from RF+camera data using contrastive learning. The framework outperformed its supervised counterpart on downstream tasks even with less training data – accepted at IEEE ICC 2022
  • ML Stack: Self-supervised learning, contrastive learning, CNNs, multi-modal fusion, and self-attention
  • Tech Stack: Python, PyTorch, Scikit-learn