Hire Jason Q. - AI Engineer - New York

Developed novel deep learning architecture for combinatorial drug discovery, driving up to 40% increase in target identification. First-author publication and poster presentation at the main track of ICML 2025 - Created internal agentic retrieval-augmented generation LLM. Leveraged agentic AI, public SQL datasets, and 3M+ pages of unstructured text to evaluate potential therapeutic indications - Identified potential drug targets for multiple diseases through exploratory data analysis and statistical modeling tools including dimensionality reduction, unsupervised clustering, and non-parametric modeling

Implemented cutting-edge natural language models to predict antigen binders for T-Cell Receptors - Developed novel statistical models in Python to predict mRNA splicing from RNA-seq data Projects: - Implemented auxiliary loss and sequential layer unfreezing in PyTorch and NumPy to improve natural language model for question answering over state-of-the-art baseline by 13%. Received Best Project Award (out of 150+ submitted projects) for Stanford CS224N: Natural Language Processing - Adapted and trained deep learning model (ResNet CNN) in PyTorch for computer vision task to identify tumors in CT brain scans. Leveraged regularization, data augmentation, and hyperparameter scans. - Sentiment analysis (using naïve bayes algorithm) in Python, NumPy to classify tweets into 3 categories (supporting, neutral, or opposing in belief of human-driven climate change) based on tweet text