Gaurav Vaishnav — Portfolio

Projects

Certified

Self-directed project

AI-Assisted Code Modernization with IBM

Junior Software Engineer · Software Engineering · October 2026

96/ 100

Based on a public IBM case study · not affiliated with IBM

Built and verified an end-to-end modernization of a legacy fleet-maintenance service: fixed the wear calculation, handled missing service readings safely, corrected the nightly report’s average and crash path, and repaired a hidden mileage-conversion bug in the helpers. Also added the missing regression test, documented the agent review in NOTES.md, and produced a breakdown-risk analysis grounded in the actual fleet history data rather than assumptions.

Graded against

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

Passed · pass mark 60/100

What stood out6

Caught the floor-division wear bug and correctly predicted the near-worn car should read about 99%
Swept the helper module and fixed the quiet km-to-miles conversion error that no test directly caught
Kept the 15000 km interval and 80% threshold unchanged while still modernizing the code style
Wrote a data-driven risk analysis that explicitly rejects the tempting but unsupported age/odometer story
Leakage-Free Feature Engineering
Honest Model Evaluation

The deliverable

View the codegaurvais / proov-km-waechter-fix

The work I submitted20 tasks

12 tasks · 2.3k characters · 8 written answers · 84 words

Division Predict70 characters
Diagnose Line130 characters
Diagnose Why136 characters
Scope Prompt66 characters
Audit Findings158 characters
Audit Why172 characters
Acceptance Checklist147 characters
Bob Trial Click74 characters
Bob Ready60 characters
Template Repo Click82 characters
Verify Report71 characters
Exec Pitch1.1k characters
Division Predict1 word
Diagnose Line10 words
Diagnose Why22 words
Scope Prompt1 word
Audit Findings1 word
Audit Why25 words
Acceptance Checklist23 words
Bob Trial Click1 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

Business Analytics Sprint

Data Science · August 2026

91/ 100

A ProoV case study · educational project, not employment

Built an end-to-end profit investigation that started with structured hypotheses and a data audit, then validated the core drivers with SQL, pivot analysis, and a Pareto-style root-cause ranking. The work culminated in a boardroom-ready recommendation set and executive summary that linked discount discipline, logistics efficiency, and raw-material cost control to a credible recovery path.

Graded against

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

Passed · pass mark 60/100

What stood out6

Built a structurally sound hypothesis set around raw materials, logistics, and discounting before analysis
Computed category margin with the correct revenue-minus-cost logic and grouped by category
Used a joined regional query to connect margin with discount behavior
Translated the analysis into a prioritized boardroom narrative with clear growth-versus-profit framing
Leakage-Free Problem Framing
Correct SQL Aggregation and Join Logic

The work I submitted20 tasks

12 tasks · 4.8k characters · 8 written answers · 248 words

Hypothesis Sheet890 characters
KPI Driver Map174 characters
Data Audit100 characters
Week1 Update648 characters
Sql Category Profit297 characters
Sql Region Join367 characters
Dashboard58 characters
Root Causes158 characters
Forecast274 characters
AI Reflection205 characters
Recommendations964 characters
Boardroom Deck651 characters
Hypothesis Sheet98 words
KPI Driver Map1 word
Data Audit4 words
Week1 Update93 words
Sql Category Profit19 words
Sql Region Join27 words
Dashboard1 word
Root Causes5 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

Data Analytics and AI

Data Science · June 2026

84/ 100

A ProoV case study · educational project, not employment

A verified, self-directed industry project — completed and passed a real industry rubric at 84/100 on ProoV.

Graded against

Exploratory Data Analysis20%
Time-Series Analysis25%
Forecasting Model Building35%
Business Communication20%

Passed · pass mark 60/100

What stood out6

Compared three forecasting approaches with MAE, RMSE, and MAPE on a sequential holdout split
Correctly identified SARIMA as the strongest model for seasonal demand forecasting
Produced meaningful regional and product/channel insights tied to retail strategy
Used decomposition, moving averages, and ACF to support time-series interpretation
Data-Driven Decision Making
Time-Series Analysis

The work I submitted20 tasks

12 tasks · 11.7k characters · Python · 8 written answers · 595 words

Setup Resource DeliveryPython · 1.0k characters
Eda Code 1 LoadPython · 1.6k characters
Eda Step 1 Load410 characters
Eda Code 1 Describe2.0k characters
Eda Step 1 Describe481 characters
Eda Code 1 Timeseries880 characters
Eda Step 1 Timeseries460 characters
Eda Code 1 Regional1.1k characters
Eda Step 1 Regional600 characters
Eda Code 1 Product1.6k characters
Ts Code 2 Resample1.2k characters
Ts Step 2 Resample432 characters
Setup Resource Delivery84 words
Eda Code 1 Load101 words
Eda Step 1 Load54 words
Eda Code 1 Describe90 words
Eda Step 1 Describe61 words
Eda Code 1 Timeseries65 words
Eda Step 1 Timeseries61 words
Eda Code 1 Regional79 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 · June 2026

86/ 100

A ProoV case study · educational project, not employment

A verified, self-directed industry project — completed and passed a real industry rubric at 86/100 on ProoV.

Graded against

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

Passed · pass mark 60/100

What stood out6

Correctly identified that Random Forest substantially outperformed Linear Regression on R² and MAE.
Highlighted the most important pricing drivers, including MPG, car age, and engine size.
Connected model performance to a concrete CPO business impact estimate and pricing recommendation.
Demonstrated useful EDA insights, including Audi's price premium over VW.
Data-Driven Decision Making
Feature Engineering

The work I submitted10 tasks

5 tasks · 6 lines · 5 written answers · 166 words

VW Sprint1 Complete56 characters
VW Sprint2 Complete61 characters
VW Sprint3 Complete59 characters
VW Business Insights1.3k characters
Jupyterlite Code2 lines
VW Sprint1 Complete1 word
VW Sprint2 Complete1 word
VW Sprint3 Complete1 word
VW Business Insights162 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