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 full legacy-fix pass for a fleet-maintenance service: corrected the wear calculation, handled missing service readings safely, fixed the nightly report’s average and crash path, repaired a hidden mileage-conversion bug in the helpers, and added the missing regression test. Also produced a data-driven breakdown-risk analysis that follows the fleet history rather than the misleading mileage assumption, with documented verification in NOTES.md.
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 integer-division wear bug and explained why it suppresses a nearly-worn car
- Caught the quiet helper-module mileage conversion bug that no test directly covered
- Wrote a substantive NOTES.md describing what the agent got wrong and how the fix was verified
- Built an analysis that explicitly rejects the misleading odometer/age assumption and uses the real predictive factors
- Leakage-Free Bug Diagnosis
- Directed AI Prompt Scoping
The work I submitted18 tasks
10 tasks · 1.4k characters · 8 written answers · 53 words
- Division Predict77 characters
- Diagnose Line130 characters
- Diagnose Why124 characters
- Scope Prompt66 characters
- Audit Findings146 characters
- Audit Why111 characters
- Bob Trial Click74 characters
- Template Repo Click152 characters
- Exec Pitch326 characters
- Bob Bonus164 characters
- Division Predict1 word
- Diagnose Line10 words
- Diagnose Why21 words
- Scope Prompt1 word
- Audit Findings1 word
- Audit Why17 words
- Bob Trial Click1 word
- 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.