Watch AI Fix a Real Bug

AI can write code in seconds.
But can it fix a REAL bug in a REAL banking system, correctly?
Let's find out. Live.
🏦 THE CODEBASE: APACHE FINERACT
Not a to-do app. Not a toy demo.
Apache Fineract is an open-source core banking platform used by financial institutions around the world to manage loans, savings, accounts and accounting.
→ Large, multi-module Java & Spring Boot codebase
→ Real banking logic: loans, repayments, interest, accounting
→ The kind of complex, legacy-scale system you work on every day
If AI can work here, it can work on your project.
🐞 THE BUG
A customer makes a loan repayment. The network is slow, the request is retried, and the repayment is posted TWICE.
Wrong balance. Broken accounting. An angry customer. And no obvious error in the logs.
In banking, a bug like this costs real money and real trust. It's exactly the kind of bug AI often gets confidently wrong.
⚔️ THE SHOWDOWN
Claude and Codex get the same bug on the same Fineract codebase. We'll fix it step by step, the way production banking teams work:
1️⃣ Understand the system: AI maps a huge, unfamiliar codebase in minutes
2️⃣ Reproduce the bug: prove the double posting before touching any code
3️⃣ Find the root cause: trace the repayment flow with evidence, not guesses
4️⃣ Write a failing test: lock the bug down so it can never return
5️⃣ Apply the smallest safe fix: no risky rewrites in financial code
6️⃣ Verify everything: run the tests and confirm balances and accounting are correct
7️⃣ AI code review: a second AI reviews the fix like a senior banking engineer
You'll also see where AI gets it WRONG on a complex codebase, and how a skilled engineer catches it before it reaches production.
💡 WHY THIS MATTERS FOR YOU
→ Freshers: see how real enterprise code is debugged, the skill interviews test
→ Developers: fix bugs in large codebases faster, without shipping AI mistakes
→ Senior devs & leads: make AI follow standards on complex, high-risk systems
→ Architects & managers: see what AI can safely do in regulated, real-world delivery
🎁 YOU'LL TAKE HOME
✅ A 7-step AI bug-fixing workflow for large enterprise codebases
✅ Ready-to-use prompts that get correct answers, not plausible ones
✅ An honest view of when to use Claude, when to use Codex, and when to trust neither
✅ A real open-source project to practise on after the session
✅ Answers to your questions in an open Q&A
Demo in Java on Fineract. The workflow works for every stack: Java, .NET, Python and more.
No slides marathon. Just real code, real bugs, real fixes.
💻 Want to follow along? Explore Apache Fineract on GitHub before the session: github.com/apache/fineract

Limited seats to keep it interactive. Register now.

━━━━━━━━━━━━━━━━━━
VENKATESH DB
Enterprise AI that ships.
━━━━━━━━━━━━━━━━━━