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 validated a full legacy-code modernization for a fleet service monitor: corrected the wear calculation, handled missing service history safely, fixed the nightly report’s average and crash path, repaired a hidden mileage-conversion bug in shared helpers, and added regression coverage. Also produced a data-driven breakdown-risk analysis that ranks cars by service-related risk rather than relying on misleading mileage or age 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
- Fixed the quiet helper bug in fleet_utils.km_to_miles, which is the strongest sign they swept beyond the obvious failures.
- Added a missing regression test for the no-reading fleet-report edge case.
- Built analyze.py around the actual signal in the data, explicitly rejecting odometer_km and age_years as predictors.
- Leakage-Free Bug Fixing
- Honest Acceptance Checking
- Data-Driven Risk Analysis
The deliverable
The work I submitted20 tasks
12 tasks · 2.1k characters · 8 written answers · 66 words
- Division Predict72 characters
- Diagnose Line130 characters
- Diagnose Why199 characters
- Scope Prompt66 characters
- Audit Findings159 characters
- Audit Why137 characters
- Bob Trial Click74 characters
- Bob Ready60 characters
- Template Repo Click82 characters
- Exec Pitch717 characters
- Bob Bonus361 characters
- Repo URL57 characters
- Division Predict1 word
- Diagnose Line10 words
- Diagnose Why31 words
- Scope Prompt1 word
- Audit Findings1 word
- Audit Why20 words
- Bob Trial Click1 word
- Bob Ready1 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.