Certified
Self-directed project
Machine Learning for Automotive
AI / ML · July 2026
92/ 100
Built an end-to-end used-car pricing workflow for VW and Audi listings: combined the datasets, explored price and mileage patterns, engineered predictive features, and trained both Linear Regression and Random Forest models. The work culminated in a business summary that translated model results into a practical recommendation for a Certified Pre-Owned purchasing dashboard.
Graded against
- Exploratory Data Analysis20%
- Feature Engineering25%
- Model Building & Evaluation35%
- Business Communication20%
Passed · pass mark 60/100
What stood out6
- Combined VW and Audi listings with a brand column and used the merged dataset consistently in analysis
- Added the key predictive features and cleaned obvious data errors before modeling
- Compared Linear Regression against Random Forest using R², MAE, and RMSE rather than relying on a single metric
- Wrote an executive summary that connects model performance to a concrete CPO purchasing recommendation
- Leakage-Free Feature Engineering
- Honest Model Evaluation
The work I submitted5 tasks
VW Sprint1 Complete
My solution
{"completed":true,"notebookName":"vw_sprint1_eda.ipynb"}VW Sprint2 Complete
My solution
{"completed":true,"notebookName":"vw_sprint2_features.ipynb"}VW Sprint3 Complete
My solution
{"completed":true,"notebookName":"vw_sprint3_models.ipynb"}VW Business Insights
My solution
{"text":"To: VP of Used Car Sales\nFrom: Serafim, Data Scientist\nSubject: Executive Summary - Machine Learning Pricing Optimization\n\nOur new Random Forest model predicts used car prices with 95% accuracy ($R^2$ = 0.95), significantly outperforming our baseline models. By upgrading to this algorithm, we reduced our average estimation error (MAE) by 45%, dropping it to just £1,438 per vehicle.\n\nThe algorithm identified the three most critical factors driving our resale pricing: fuel efficiency (MPG), vehicle age, and engine size.\n\nFinancial Impact: At our current scale of 100,000 used cars sold annually, an average baseline error of ~£1,500 represents a massive £150 million in total portfolio mispricing risk. Relying on manual pricing or less accurate methods exposes us to significant margin erosion.\n\nRecommendation: I strongly recommend fully integrating this predictive model directly into the Certified Pre-Owned (CPO) purchasing dashboard. This operational shift will empower our acquisition team to instantly flag underpriced incoming inventory, standardize our buying decisions across all branches, and systematically secure higher-margin vehicles faster than the competition.","charCount":1182,"selfChecks":[{"label":"Model comparison","checked":false},{"label":"Top features identified","checked":false},{"label":"Financial Impact","checked":false},{"label":"Clear recommendation","checked":false}]}Jupyterlite Code
My solution
base64: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
…