Swastik Roy — Portfolio

Projekte

Zertifiziert

Selbstgesteuertes Projekt

Machine Learning for Automotive

KI / ML · September 2026

68/ 100

Eine ProoV-Fallstudie · Lernprojekt, kein Arbeitsverhältnis

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, trained and compared Linear Regression and Random Forest models, and translated the results into a Certified Pre-Owned pricing recommendation. The work shows practical data analysis, model evaluation, and business-facing communication.

Bewertet nach

Exploratory Data Analysis20%
Feature Engineering25%
Model Building & Evaluation35%
Business Communication20%

Bestanden · Bestehensgrenze 60/100

Was herausstach6

Combined VW and Audi listings into a single analysis frame with a brand indicator and explored distributions, correlations, and brand-level price differences.
Built a naive mileage-only baseline and then moved to engineered features and model comparison, showing an understanding of why richer predictors matter.
Identified the main valuation drivers as registration year, engine size, and mileage, and translated model output into a CPO pricing recommendation.
Exploratory Data Analysis
Feature Engineering
Model Comparison

Meine eingereichte Arbeit10 Aufgaben

5 Aufgaben · 6 Zeilen · 5 Textantworten · 147 Wörter

VW Sprint1 Complete56 Zeichen
VW Sprint2 Complete61 Zeichen
VW Sprint3 Complete59 Zeichen
VW Business Insights1.3k Zeichen
Jupyterlite Code2 Zeilen
VW Sprint1 Complete1 Wort
VW Sprint2 Complete1 Wort
VW Sprint3 Complete1 Wort
VW Business Insights143 Wörter
Jupyterlite Code1 Wort

Bewusst zusammengefasst — der eingereichte Code und die Texte bleiben privat, damit diese Seite nicht als Lösungsvorlage dient. Die vollständige Einreichung steht hinter dem verifizierten Zertifikat.

Verifiziertes ZertifikatFälschungssicher · ausgestellt von ProoV

Zertifiziert

Selbstgesteuertes Projekt

AI-Assisted Code Modernization with IBM

Junior Software Engineer · Softwareentwicklung · August 2026

92/ 100

Basierend auf einer öffentlichen Fallstudie von IBM · nicht mit IBM verbunden

Built and verified a full legacy-code modernization for KM-Waechter: fixed the service-wear calculation, handled missing service readings safely, repaired the nightly report’s average and crash path, corrected a hidden mileage-conversion bug in shared helpers, and added the missing regression test. Also delivered a data-driven breakdown-risk analysis that ranks cars by recent service wear and workload rather than relying on misleading mileage or age assumptions.

Bewertet nach

Fix and extend the repo with the agent35%
Diagnose the bug20%
Direct and audit the AI agent25%
Explain it like an engineer20%

Bestanden · Bestehensgrenze 60/100

Was herausstach6

Correctly identified and fixed the wear calculation so a 14,900 km car is flagged instead of being floored to 0%
Swept beyond the obvious tests and fixed the quiet fleet_utils.km_to_miles conversion bug that verify.py specifically checks
Added the missing no-crash report test and documented the agent review process in NOTES.md
Produced an analysis that explicitly rejects the misleading mileage/age assumption and focuses on km_since_service, avg_daily_km, and load_factor
Leakage-Free Bug Diagnosis
Directed AI Code Review

Das Ergebnis

Zum Coderoyswas389 / proov-km-waechter-fix

Meine eingereichte Arbeit20 Aufgaben

12 Aufgaben · 1.7k Zeichen · 8 Textantworten · 108 Wörter

Division Predict70 Zeichen
Diagnose Line130 Zeichen
Diagnose Why167 Zeichen
Scope Prompt66 Zeichen
Audit Findings158 Zeichen
Audit Why139 Zeichen
Acceptance Checklist217 Zeichen
Bob Trial Click136 Zeichen
Bob Ready60 Zeichen
Template Repo Click82 Zeichen
Verify Report71 Zeichen
Exec Pitch362 Zeichen
Division Predict1 Wort
Diagnose Line10 Wörter
Diagnose Why29 Wörter
Scope Prompt1 Wort
Audit Findings1 Wort
Audit Why24 Wörter
Acceptance Checklist41 Wörter
Bob Trial Click1 Wort

Verifiziertes ZertifikatFälschungssicher · ausgestellt von ProoV

Zertifiziert

Selbstgesteuertes Projekt

Fundamentals of Cybersecurity

Cybersicherheit · August 2026

78/ 100

Eine ProoV-Fallstudie · Lernprojekt, kein Arbeitsverhältnis

Built a beginner SOC incident response workflow end to end: triaged a phishing email, identified anomalous authentication activity including impossible travel, selected the relevant IOCs, and sequenced containment to cut access before taking broader action. Also produced a structured breach notification and incident report that connected the phishing entry point to the account compromise and data exposure.

Bewertet nach

Threat Triage & Detection20%
Log Analysis & Investigation30%
Incident Response & Containment20%
Compliance & Breach Notification15%
Executive Communication & Lessons15%

Bestanden · Bestehensgrenze 60/100

Was herausstach6

Correctly flagged all four phishing red flags: sender, link, attachment, and urgency without false positives
Identified the impossible-travel sign-in and caught five anomalous auth rows with no benign rows blocked
Chose a sound containment order that cut access first and avoided destructive actions
Wrote a structured incident report with timeline, root cause, impact, and recommendations tied to the breach
Leakage-Free Threat Triage
Anomalous Login Investigation

Meine eingereichte Arbeit18 Aufgaben

10 Aufgaben · 4.5k Zeichen · 8 Textantworten · 300 Wörter

Phishing Triage186 Zeichen
Killchain Map232 Zeichen
Log Triage190 Zeichen
Ioc Extraction176 Zeichen
Containment Plan373 Zeichen
Breach Notification864 Zeichen
Incident Report1.0k Zeichen
Recommendations649 Zeichen
Real Case Analysis498 Zeichen
Teach Back294 Zeichen
Phishing Triage8 Wörter
Killchain Map8 Wörter
Log Triage9 Wörter
Ioc Extraction8 Wörter
Containment Plan14 Wörter
Breach Notification114 Wörter
Incident Report126 Wörter
Recommendations13 Wörter

Verifiziertes ZertifikatFälschungssicher · ausgestellt von ProoV