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 gas-turbine TEY predictor using the eight pre-combustion sensors, with CO and NOx excluded because they are downstream emissions. Validated on a chronological holdout, compared Linear Regression against a Random Forest, and reported honest held-out performance (about 1.5 MWh MAE, ~0.98 R2) alongside residual analysis and a practical monitoring plan.
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 documented why
- Used a time-based split instead of a random shuffle, preserving the integrity of the evaluation
- Reported held-out MAE, RMSE, and R2, then read the residuals for bias and heteroscedasticity
- Wrote a model card that names the plant-specific limits, silent-drift risk, and a concrete retraining trigger
- Leakage-Free Feature Selection
- Chronological Holdout Validation
The work I submitted20 tasks
12 tasks · 13 lines · Python, JavaScript · 8 written answers · 431 words
- Problem Frame238 characters
- Leakage Flags113 characters
- Train Validate BaselinePython · 1.4k characters
- Train Validate124 characters
- Train Validate Forest944 characters
- Tune Evaluate SetupPython · 1.5k characters
- Tune Evaluate ResidualsPython · 2 lines
- Tune Evaluate130 characters
- Residual Read256 characters
- Drift Monitor92 characters
- Model Card1.1k characters
- Teach BackJavaScript · 736 characters
- Problem Frame20 words
- Leakage Flags1 word
- Train Validate Baseline141 words
- Train Validate1 word
- Train Validate Forest77 words
- 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.