K Nithin Kumar Goud — Portfolio

Projects

Certified

Self-directed project

Predictive Maintenance: Industrial ML for Fault Detection

Condition-Monitoring Engineer · AI / ML · August 2026

96/ 100

A ProoV case study · educational project, not employment

Built an end-to-end rolling-bearing fault detector from vibration data: visualized a faulty snapshot, engineered RMS/kurtosis/crest-factor and defect-frequency envelope features, localized the fault to the outer race via BPFO, and trained a tiny depth-capped decision tree on training bearings only. Evaluated on a held-out set with catch-rate and false-alarm metrics, then justified a 0.30 alarm threshold using the candidate’s own cost numbers in a maintenance memo.

Graded against

Feature engineering (signal to honest numbers)25%
Held-out evaluation (honest measurement)25%
Cost-based threshold decision + memo25%
Compact model + embedded-budget justification15%
Plain-language explanation10%

Passed · pass mark 60/100

What stood out6

Used the provided envelope_spectrum() helper and correctly found BPFO as the dominant defect band, mapping it to the outer race.
Reported a compact decision tree with stated depth 1 and 3 nodes, explicitly noting it was trained on training bearings only and within the embedded budget.
Made the threshold decision with concrete operational costs (€20,000 missed failure vs €500 false alarm) and tied the memo to the observed 100% catch-rate and 42% false-alarm rate.
Leakage-Free Feature Engineering
Honest Held-Out Evaluation
Physics-Based Fault Localization

The work I submitted19 tasks

11 tasks · 12 lines · Python · 8 written answers · 401 words

Initial Prediction62 characters
Load Plot Signal CodePython · 1.7k characters
Example Rms WorkedPython · 697 characters
Time Features143 characters
Freq FeaturesPython · 935 characters
Fault Location239 characters
Train Classifier363 characters
Evaluation Heldout288 characters
Threshold Decision334 characters
Decision Memo1.2k characters
Jupyterlite Code2 lines
Initial Prediction9 words
Load Plot Signal Code163 words
Example Rms Worked72 words
Time Features10 words
Freq Features78 words
Fault Location14 words
Train Classifier31 words
Evaluation Heldout24 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

Driving Supply Chain Sustainability via Data Governance

Product Compliance Specialist · General · August 2026

84/ 100

A ProoV case study · educational project, not employment

Built a three-part battery-passport compliance package for a single cell using the PowerCo Salzgitter case as reference. Classified regulated fields into known, computable, and gap buckets; designed public, legitimate-interest, and restricted passport views; and traced unresolved gaps to the supplier tier that owns them. Also computed the cell’s carbon footprint from code and used the result to identify raw materials as the dominant contributor.

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

Defended unsettled passport fields with explicit criteria that would change the classification
Computed the carbon footprint from code and correctly identified raw materials as the dominant stage
Mapped remaining gaps to the supplier tier that actually owns the data
Drafted supplier requests with field names, regulatory basis, deadlines, and batch-level specificity
Leakage-Free Regulatory Classification
Traceable Judgment Under Ambiguity

The work I submitted2 tasks

1 task · 2 lines · 1 written answer · 1 word

Jupyterlite Code2 lines
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

AI-Assisted Code Modernization with IBM

Junior Software Engineer · Software Engineering · August 2026

94/ 100

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

Built and verified a full legacy-service repair for Vossberg Mobility: fixed the wear calculation so nearly-worn cars are flagged, handled missing service readings safely, corrected the fleet report’s average and crash path, and preserved the 15,000 km / 80% business rules. Also added the missing regression test, modernized helper code, and produced a data-driven breakdown-risk analysis that follows the fleet history rather than the naive mileage assumption.

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

Correctly identified the floor-division bug and explained why 14,900/15,000 became 0% wear
Caught and reverted the agent’s unauthorized threshold change from 80% to 85%
Added the missing regression test for the nearly-worn car and documented the acceptance checks
Wrote NOTES.md with concrete, self-audited findings, including the helper-module mileage conversion bug
Leakage-Free Bug Diagnosis
Verified AI Output Auditing

The deliverable

View the codenithingoud78 / km-waechter-fix

The work I submitted20 tasks

12 tasks · 2.1k characters · 8 written answers · 76 words

Bob Trial Click198 characters
Bob Ready60 characters
Diagnose Line130 characters
Diagnose Why163 characters
Scope Prompt66 characters
Audit Findings158 characters
Audit Why119 characters
Acceptance Checklist131 characters
Template Repo Click82 characters
Exec Pitch709 characters
Bob Bonus229 characters
Repo URL51 characters
Bob Trial Click1 word
Bob Ready1 word
Diagnose Line10 words
Diagnose Why24 words
Scope Prompt1 word
Audit Findings1 word
Audit Why19 words
Acceptance Checklist19 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

Frontend Engineering with Prediction Game

Junior Front-End Engineer · Software Engineering · July 2026

84/ 100

A ProoV case study · educational project, not employment

Built a complete World Cup prediction game end to end: rendered the tournament fixtures as interactive pick cards, scored picks with the exact 5/2/0 rubric, derived team attack/defense strengths from real historical results, and used those strengths to generate bot predictions. Also implemented a persistent ranked leaderboard and deployed the finished experience to a live Netlify URL.

Graded against

Scoring engine correctness30%
Prediction model genuinely from real data25%
Game completeness and leaderboard soundness25%
Shipped and shareable: live deployed URL20%

Passed · pass mark 60/100

What stood out6

Rendered the fixture board with one card per match and two numeric score inputs per card.
Implemented scorePick with correct handling for exact scores, same-result different-score cases, and draws.
Computed team strengths from a real-results sample and used those strengths inside botPredict rather than hard-coding outcomes.
Shipped a live https deployment on Netlify.
Leakage-Free Feature Engineering
Honest Model Evaluation

The work I submitted18 tasks

10 tasks · 12 lines · JavaScript · 8 written answers · 664 words

Lab Orientation677 characters
Example Render FixturesJavaScript · 2.0k characters
Act1 Fixtures FormJavaScript · 3.6k characters
Act2 Scoring FunctionJavaScript · 2.3k characters
Act3 Compute Strength108 characters
Act3 House Bot108 characters
Act4 LeaderboardJavaScript · 3.2k characters
Act5 Assemble2 lines
Final Submission95 characters
Jupyterlite Code2 lines
Lab Orientation66 words
Example Render Fixtures151 words
Act1 Fixtures Form116 words
Act2 Scoring Function113 words
Act3 Compute Strength1 word
Act3 House Bot1 word
Act4 Leaderboard108 words
Act5 Assemble108 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

Amazon Robotics - Black-Friday Fleet Coordination

Software Engineering · July 2026

98/ 100

A ProoV case study · educational project, not employment

Built an end-to-end Amazon Robotics fleet-coordination solution for Black Friday surge conditions: framed the bottleneck as aisle congestion and dock contention, implemented single-agent A* with an admissible Manhattan heuristic, extended it to centralized cooperative planning with vertex and edge reservations, and validated a realistic operating policy in simulation. The final pitch tied zero-collision execution and sustained throughput to a concrete pilot ask for one FC zone during a named surge event.

Graded against

Problem framing & operational stakes20%
Algorithmic depth - A* and cooperative coordination30%
Coordination policy realism - deadlock, charging, idle25%
Throughput trade-off analysis15%
Executive pitch to leadership10%

Passed · pass mark 55/100

What stood out6

Implemented A* with an admissible Manhattan heuristic on a 4-connected grid and showed it outperforming Dijkstra in node expansions.
Built a cooperative reservation table that books both vertices and directed edges, preventing same-cell and head-on swap collisions.
Chose a realistic fleet policy: centralized coordination, priority-watchdog deadlock handling, staggered charging, and dispersed perimeter idle parking.
Used simulator outputs to identify a concave throughput curve and locked a specific sweet spot fleet size of 36 robots.
Leakage-Free Multi-Agent Coordination
Honest Simulation-Based Evaluation

The work I submitted20 tasks

12 tasks · 7.9k characters · Python · 8 written answers · 376 words

Amazon Robotics Context459 characters
Grid World Theory119 characters
State Representation Task ValidatorPython · 1.7k characters
State Representation Task341 characters
Phase1 Recap348 characters
Astar Theory124 characters
Astar Implementation Cell3.7k characters
Astar Solution Correct4 characters
Single Unit Challenge465 characters
Phase2 Recap374 characters
Multi Agent Problem111 characters
Reservation Table Theory180 characters
Amazon Robotics Context37 words
Grid World Theory1 word
State Representation Task Validator129 words
State Representation Task1 word
Phase1 Recap23 words
Astar Theory1 word
Astar Implementation Cell171 words
Single Unit Challenge13 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

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%
Feature Engineering25%
Model Building & Evaluation35%
Business Communication20%

Passed · pass mark 60/100

What stood out6

Correctly compared Linear Regression and Random Forest using R², MAE, and RMSE
Identified vehicle age, mileage, and engine size as the top pricing drivers
Translated model performance into a clear CPO business recommendation
Included a quantified financial impact statement to support decision-making
Exploratory Data Analysis
Feature Engineering

The work I submitted10 tasks

5 tasks · 6 lines · 5 written answers · 186 words

VW Sprint1 Complete56 characters
VW Sprint2 Complete61 characters
VW Sprint3 Complete59 characters
VW Business Insights1.6k characters
Jupyterlite Code2 lines
VW Sprint1 Complete1 word
VW Sprint2 Complete1 word
VW Sprint3 Complete1 word
VW Business Insights182 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