Self-directed project
Energy and Renewable: Machine Learning for the Power Grid
Junior Machine Learning Engineer · AI / ML · July 2026
A ProoV case study · educational project, not employment
Built a leakage-aware turbine yield regressor end to end: framed TEY as the target, excluded downstream CO/NOx emissions, trained a LinearRegression baseline and a RandomForest model, and validated on a genuine time-based holdout. The final model was evaluated honestly with held-out MAE, RMSE, and R2, and the accompanying model card documented intended use, limitations, and monitoring considerations.
Graded against
- Build and validate the model correctly35%
- Optimize and evaluate honestly25%
- Write the Model Card (clarity + honesty)25%
- Explain it to a non-expert15%
Passed · pass mark 60/100
What stood out6
- Correctly excluded CO and NOx as downstream leakage features and used the 8 legitimate sensor inputs.
- Used an earlier-period train / later-period test split instead of a random shuffle, which is the right validation pattern for time-ordered data.
- Compared a LinearRegression baseline against a RandomForest and showed a meaningful held-out MAE improvement.
- Included held-out MAE, RMSE, R2, and a residual-read statement that avoids overclaiming generalization.
- Leakage-Free Feature Selection
- Time-Based Holdout Evaluation
The work I submitted19 tasks
11 tasks · 7.2k characters · Python, JavaScript · 8 written answers · 313 words
- Problem Frame143 characters
- Leakage Flags113 characters
- Train Validate BaselinePython · 1.8k characters
- Train Validate Forest965 characters
- Train Validate123 characters
- Tune Evaluate SetupPython · 1.5k characters
- Tune Evaluate89 characters
- Residual Read285 characters
- Drift Monitor93 characters
- Model Card1.3k characters
- Teach BackJavaScript · 777 characters
- Problem Frame10 words
- Leakage Flags1 word
- Train Validate Baseline88 words
- Train Validate Forest63 words
- Train Validate1 word
- Tune Evaluate Setup110 words
- Tune Evaluate1 word
- Residual Read39 words
Summarised on purpose — the submitted code and writing stay private so this page cannot be reused as an answer key. The full submission sits behind the verified certificate.