Self-directed project
Energy and Renewable: Machine Learning for the Power Grid
Junior Machine Learning Engineer · AI / ML · September 2026
A ProoV case study · educational project, not employment
Built a leakage-aware turbine yield predictor 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 locked-away later-year holdout. The final model achieved about 1.5 MWh MAE and 0.98 R² on unseen data, with a model card that clearly states intended use, limitations, and monitoring guidance.
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 emissions that would not be available at prediction time
- Used an earlier-years/later-years split instead of a random shuffle, preserving temporal validity
- Compared a linear baseline to a tuned RandomForest and judged improvement on held-out MAE
- Reported held-out MAE, RMSE, and R² and explicitly cautioned against overgeneralizing beyond this plant
- Leakage-Free Feature Selection
- Time-Based Holdout Evaluation
The work I submitted19 tasks
11 tasks · 6.9k characters · Python, JavaScript · 8 written answers · 386 words
- Problem Frame238 characters
- Leakage Flags113 characters
- Train Validate BaselinePython · 1.4k characters
- Train Validate Forest943 characters
- Train Validate124 characters
- Tune Evaluate SetupPython · 1.5k characters
- Tune Evaluate91 characters
- Residual Read254 characters
- Drift Monitor94 characters
- Model Card1.4k characters
- Teach BackJavaScript · 772 characters
- Problem Frame23 words
- Leakage Flags1 word
- Train Validate Baseline141 words
- Train Validate Forest76 words
- Train Validate1 word
- Tune Evaluate Setup110 words
- Tune Evaluate1 word
- Residual Read33 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.