Self-directed project
AI-Assisted Code Modernization with IBM
Junior Software Engineer · Software Engineering · August 2026
Based on a public IBM case study · not affiliated with IBM
Built and verified a modernized KM-Waechter fix end to end: corrected the wear calculation with true division, handled missing service readings safely, repaired the nightly fleet report, added the missing regression test, and documented the review in NOTES.md. Also delivered a data-driven breakdown-risk analysis that ranks cars using the factors the dataset actually supports, not the misleading 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 it turned a nearly worn car into 0% wear
- Caught the agent’s bad fix of changing the threshold instead of fixing the operator
- Produced an analysis that follows the dataset’s real signals rather than the misleading mileage/age assumption
- Leakage-Free Bug Diagnosis
- Verified AI Output Auditing
- Data-Driven Risk Analysis
The deliverable
The work I submitted20 tasks
12 tasks · 2.2k characters · 8 written answers · 92 words
- Bob Trial Click74 characters
- Bob Ready60 characters
- Diagnose Line130 characters
- Diagnose Why170 characters
- Scope Prompt66 characters
- Audit Findings158 characters
- Audit Why133 characters
- Acceptance Checklist137 characters
- Template Repo Click82 characters
- Exec Pitch661 characters
- Bob Bonus449 characters
- Repo URL61 characters
- Bob Trial Click1 word
- Bob Ready1 word
- Diagnose Line10 words
- Diagnose Why32 words
- Scope Prompt1 word
- Audit Findings1 word
- Audit Why23 words
- Acceptance Checklist23 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.