Self-directed project
Energy and Renewable: Machine Learning for the Power Grid
Junior Machine Learning Engineer · AI / ML · August 2026
A ProoV case study · educational project, not employment
Built an end-to-end TEY prediction workflow for a gas-turbine plant using leakage-free sensor features and a time-based train/test split. Compared a LinearRegression baseline with a RandomForest model, selected the stronger tree ensemble on held-out performance, and documented the result with honest MAE/RMSE/R² plus a monitoring plan for drift and retraining.
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
- Dropped CO and NOx explicitly because they are downstream emissions and unavailable at prediction time
- Used a locked later-year test split instead of a random shuffle, preserving time order
- Compared a LinearRegression baseline against a tuned RandomForest and showed a clear MAE improvement
- Included residual analysis and a concrete retraining trigger tied to rolling MAE
- Leakage-Free Feature Selection
- Honest Held-Out Evaluation
The work I submitted19 tasks
11 tasks · 12 lines · Python · 8 written answers · 401 words
- Problem Frame234 characters
- Leakage Flags113 characters
- Train Validate BaselinePython · 1.3k characters
- Train Validate Forest900 characters
- Train Validate124 characters
- Tune Evaluate SetupPython · 1.5k characters
- Tune Evaluate ResidualsPython · 2 lines
- Tune Evaluate130 characters
- Residual Read260 characters
- Drift Monitor91 characters
- Model Card1.2k characters
- Problem Frame18 words
- Leakage Flags1 word
- Train Validate Baseline131 words
- Train Validate Forest59 words
- Train Validate1 word
- Tune Evaluate Setup110 words
- Tune Evaluate Residuals80 words
- Tune Evaluate1 word
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.