AI/ML&Software Engineer (Model evaluation and RL/policy experimentation)
Don't disclose
Built and ran systematic model evaluation workflows for forecasting and reinforcement learning experiments, comparing multiple neural architectures and training policies. Defined validation metrics (e.g., RMSE) and success rates, then performed ablation-style investigations to understand contributors to performance. Used experimental results to guide model/policy selection and iteration across repeated runs. • Benchmarked GRU/LSTM/TCN forecasting architectures with validation RMSE reporting • Evaluated reinforcement learning policies (PPO) using task-success metrics against baselines • Conducted systematic comparison and ablation studies to quantify design impacts • Logged and validated outcomes to support model selection decisions
2026 - 2026