Anthropic is one of a handful of organizations actually shipping the kinds of agentic systems I build day-to-day at Tempus AI. I'd be most useful on the parts of the stack where production reliability meets ML — agent orchestration, eval infrastructure, retrieval, deployment.
Hand-picked because they map most directly to the work at Anthropic.
Deep learning model using PyTorch to predict branch overloads in electric power grids, implementing neural networks for critical infrastructure monitoring.
End-to-end ML pipeline for security threat detection including data preprocessing, feature engineering, model training, and evaluation on real-world security datasets.
Reinforcement learning implementation using Q-learning and Dyna-Q for automated trading strategy development with experience replay and model-based planning.
Collection of algorithmic coding challenges solved in Python and C++, covering data structures, string manipulation, and SQL/Pandas data analysis across easy to hard difficulty levels.
© 2026 Maxwell Vaglica. All rights reserved.