Cade Cermak

About

Graduate student taking advanced coursework in NLP, generative image modeling, and computer vision; built a PPO reinforcement-learning agent for a Super Mario platformer with careful reward shaping and metric-driven evaluation. In a recent internship designing credit risk models, created a 10-method feature-selection voting framework and achieved ~10% out-of-sample performance improvement while reducing features to mitigate overfitting.

Experience

Education

Key Strengths

• Built a PPO RL agent for a Super Mario platformer with attention to state representation and reward shaping
• Strong reward-function balancing mindset (optimize coins/enemy defeats without increasing death risk)
• Defines clear evaluation metrics for ML agents (level completion, coins, enemies, time)
• High-impact modeling decision: multi-method feature-selection voting table with threshold search
• Improved credit risk model performance by ~10% on out-of-sample tests while reducing feature count to limit overfitting
• Able to harmonize datasets across diverse feature-selection method requirements (numeric/categorical/null constraints)