Certified
Self-directed project
Energy and Renewable: Machine Learning for the Power Grid
Junior Machine Learning Engineer · AI / ML · August 2026
92/ 100
A ProoV case study · educational project, not employment
Built an end-to-end TEY predictor on the UCI gas-turbine data with leakage-free features, a chronological hold-out split, and honest evaluation. Compared a LinearRegression baseline to a RandomForest, reported held-out MAE/RMSE/R², and documented the model’s limits, including high-load underprediction and drift monitoring.
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 dropped CO and NOx as downstream leakage features and explained why they are unavailable at prediction time.
- Used a genuine time-based split (2011–2013 train, 2014–2015 test) instead of a random shuffle.
- Compared a linear baseline to a tuned RandomForest and judged performance on held-out data.
- Read the residuals and identified a load-dependent bias/heteroscedasticity pattern rather than relying only on headline metrics.
- Leakage-Free Feature Selection
- Chronological Hold-Out Validation
The work I submitted20 tasks
12 tasks · 13 lines · Python, JavaScript · 8 written answers · 418 words
- Problem Frame144 characters
- Leakage Flags114 characters
- Train Validate BaselinePython · 1.4k characters
- Train Validate Forest944 characters
- Train Validate124 characters
- Tune Evaluate SetupPython · 1.5k characters
- Tune Evaluate ResidualsPython · 2 lines
- Tune Evaluate130 characters
- Residual Read316 characters
- Drift Monitor93 characters
- Model Card2.6k characters
- Teach BackJavaScript · 1.4k characters
- Problem Frame7 words
- Leakage Flags1 word
- Train Validate Baseline141 words
- Train Validate Forest77 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.