What Nandigam demonstrated

Built a compact bearing-fault detector end to end: visualized a faulty vibration snapshot, identified kurtosis as the strongest impulsive fault feature, trained a depth-capped decision tree on training bearings only, and evaluated it on a held-out set with reported catch-rate and false-alarm rate. The candidate also defended an alarm threshold with explicit cost trade-offs and tied the fault call to the dominant BPFO peak in a maintenance memo.

• Leakage-Free Feature Engineering
• Honest Held-Out Evaluation
• Compact Interpretable Modeling
• Cost-Based Decision Making