Self-directed project
AI-Assisted Code Modernization with IBM
Junior Software Engineer · Software Engineering · September 2026
Based on a public IBM case study · not affiliated with IBM
Built a complete modernization pass for a legacy fleet-maintenance service: fixed the wear calculation, handled missing service readings safely, corrected the nightly report, and repaired a hidden mileage-conversion bug in the helpers. Also added a data-driven breakdown-risk analysis and verified the work with a full acceptance checklist, showing disciplined AI-agent direction and review.
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 and fixed the hidden km-to-miles conversion bug that no test directly covered
- Kept the 15,000 km interval and 80% threshold unchanged while fixing the wear logic
- Wrote a data-driven breakdown analysis that explicitly rejects odometer_km and age_years as predictors
- Documented the agent’s mistakes and the acceptance checks in NOTES.md
- Leakage-Free Feature Engineering
- Honest Model Evaluation
The deliverable
The work I submitted20 tasks
12 tasks · 2.8k characters · 8 written answers · 192 words
- Division Predict73 characters
- Diagnose Line130 characters
- Diagnose Why559 characters
- Scope Prompt66 characters
- Audit Findings158 characters
- Acceptance Checklist250 characters
- Audit Why370 characters
- Template Repo Click82 characters
- Bob Trial Click266 characters
- Bob Ready60 characters
- Verify Report71 characters
- Exec Pitch672 characters
- Division Predict1 word
- Diagnose Line10 words
- Diagnose Why93 words
- Scope Prompt1 word
- Audit Findings1 word
- Acceptance Checklist33 words
- Audit Why52 words
- Template Repo 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.