Sam Dcruze — Portfolio

Projects

Certified

Self-directed project

Supply Chain Optimization with SAP Business AI

Business AI Associate · AI / ML · September 2026

88/ 100

A ProoV case study · educational project, not employment

Built a credible supply-chain rescue plan for a fictional SAP manufacturer by identifying the true bottleneck, quantifying the production risk at €120,000 per idle day, and designing a Joule-style copilot that surfaces the single best next action from live operational signals. The plan includes explicit human-in-the-loop guardrails, a defensible €235,000 payback estimate, and a board-ready pilot ask with a clear success metric.

Graded against

Problem framing & business sense30%
Business-AI solution design & guardrails30%
ROI and decision quality25%
Executive communication15%

Passed · pass mark 60/100

What stood out6

Correctly ignored the decoy port delay and focused on the true constraint: the single-sourced critical component with ~2 weeks of cover.
Defined a practical copilot that reads live supplier health, inventory cover, open POs, and port status, then drafts a PO or reroute for human approval.
Set a clear never-do boundary: the system must not autonomously onboard an unqualified supplier or commit purchases without sign-off.
Delivered a tight board ask with scope, staffing, timeframe, and success metric: a 2-person, 60-day pilot targeting detection-to-mitigation under 1 day.
Leakage-Free Problem Diagnosis
Human-in-the-Loop Copilot Design

The work I submitted19 tasks

11 tasks · 7.9k characters · 8 written answers · 578 words

SAP Brand Sentiment Pre181 characters
Trust Pledge135 characters
Diagnose Break300 characters
Problem Statement1.0k characters
Copilot Spec926 characters
Guardrails Sort1.7k characters
ROI Payback587 characters
India Transfer629 characters
Board Pitch1.6k characters
Teach Back660 characters
SAP Brand Sentiment Post188 characters
SAP Brand Sentiment Pre14 words
Trust Pledge20 words
Diagnose Break23 words
Problem Statement143 words
Copilot Spec126 words
Guardrails Sort127 words
ROI Payback53 words
India Transfer72 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.

Verified certificateTamper-proof · issued by ProoV

Certified

Self-directed project

Machine Learning for Automotive

AI / ML · September 2026

84/ 100

A ProoV case study · educational project, not employment

Built an end-to-end used-car pricing analysis for VW and Audi listings: combined the datasets, performed EDA on price, mileage, and brand differences, engineered predictive features, and compared Linear Regression with Random Forest using R² and MAE. The work culminated in a clear CPO-oriented recommendation to use the stronger model for standardized trade-in valuation and margin protection.

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 explored distributions, correlations, and brand-level price differences.
Identified the core pricing drivers in the business summary, including car age, mileage, and engine size.
Compared Linear Regression against Random Forest with explicit R² and MAE figures and tied the result to a CPO recommendation.
Exploratory Data Analysis
Feature Engineering
Model Comparison and Evaluation

The work I submitted10 tasks

5 tasks · 6 lines · 5 written answers · 141 words

VW Sprint1 Complete56 characters
VW Sprint2 Complete61 characters
VW Sprint3 Complete59 characters
VW Business Insights1.2k characters
Jupyterlite Code2 lines
VW Sprint1 Complete1 word
VW Sprint2 Complete1 word
VW Sprint3 Complete1 word
VW Business Insights137 words
Jupyterlite Code1 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.

Verified certificateTamper-proof · issued by ProoV

Certified

Self-directed project

Energy and Renewable: Machine Learning for the Power Grid

Junior Machine Learning Engineer · AI / ML · September 2026

92/ 100

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.

Verified certificateTamper-proof · issued by ProoV

Certified

Self-directed project

Driving Supply Chain Sustainability via Data Governance

Product Compliance Specialist · General · September 2026

84/ 100

A ProoV case study · educational project, not employment

Built a complete battery-passport compliance package for a single cell: classified regulated fields as known/compute/gap, designed public/legitimate-interest/restricted passport views, and traced every remaining gap to the supplier tier that owns it. Also computed the cell’s carbon footprint from code, identified raw materials as the dominant emissions source, and drafted actionable supplier requests to close the highest-priority data gaps.

Graded against

Supply-chain gap attribution and supplier requests30%
Judgment under regulatory ambiguity20%
Access-tier design judgment15%
Carbon-footprint methodology and calculation15%
Passport data-model completeness and traceability10%
Internal consistency across the package10%

Passed · pass mark 60/100

What stood out6

Computed the carbon footprint in code and reported the stage breakdown and total (74.05 kg CO2e/kWh) rather than asserting a number.
Correctly identified raw materials as the dominant emissions stage and drew the right operational conclusion about upstream sourcing leverage.
Built a coherent access-tier split that distinguishes public passport fields from recycler-useful legitimate-interest data and restricted due-diligence records.
Drafted supplier requests with named fields, regulatory basis, deadlines, and requested formats, making them actionable for real contacts.
Leakage-Free Regulatory Classification
Honest Model Evaluation

The work I submitted16 tasks

8 tasks · 8.0k characters · Python · 8 written answers · 582 words

Passport Field Defences2.3k characters
Passport Field Classification659 characters
Carbon Footprint CalculationPython · 1.3k characters
Access Tier Design918 characters
Supply Chain Gap Audit316 characters
Supplier Data Request1.3k characters
Passport Verdict Reveal490 characters
Final Submission809 characters
Passport Field Defences160 words
Passport Field Classification1 word
Carbon Footprint Calculation118 words
Access Tier Design1 word
Supply Chain Gap Audit1 word
Supplier Data Request142 words
Passport Verdict Reveal56 words
Final Submission103 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.

Verified certificateTamper-proof · issued by ProoV

Certified

Self-directed project

Business Analytics Sprint

Data Science · September 2026

92/ 100

A ProoV case study · educational project, not employment

Built an end-to-end profit-paradox analysis: framed hypotheses and KPI drivers, audited the data, wrote correct SQL for category and regional margin analysis, isolated the main profit drivers, and translated the findings into a board-ready recovery plan. The work is especially strong in connecting technical analysis to executive decisions, with a clear forecast and prioritized actions focused on margin repair.

Graded against

Problem Framing & Hypotheses10%
Data Analysis & SQL30%
Root-Cause & Forecasting25%
Dashboard & Visualization15%
Executive Communication & Strategy20%

Passed · pass mark 60/100

What stood out6

Wrote both SQL queries with the correct margin formula and appropriate GROUP BY/JOIN logic
Correctly identified the main profit drivers as raw-material inflation, logistics, and discounting
Produced a coherent forecast and tied it to prioritized recovery actions
Communicated the story clearly across memo, deck, stakeholder Q&A, and executive summary
Structured Hypothesis Formation
Leakage-Free SQL and Margin Analysis

The work I submitted20 tasks

12 tasks · 5.9k characters · 8 written answers · 301 words

Hypothesis Sheet1.1k characters
KPI Driver Map174 characters
Data Audit100 characters
Week1 Update693 characters
Sql Category Profit354 characters
Sql Region Join491 characters
Root Causes158 characters
Forecast276 characters
AI Reflection617 characters
Recommendations693 characters
Boardroom Deck676 characters
Stakeholder Qa550 characters
Hypothesis Sheet129 words
KPI Driver Map1 word
Data Audit4 words
Week1 Update91 words
Sql Category Profit27 words
Sql Region Join38 words
Root Causes5 words
Forecast6 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.

Verified certificateTamper-proof · issued by ProoV