AI Automation

AI Support Automation: Resolution Time 2.5 Days → 6 Hours

LLM tier-1 assistant in a regulated FinTech platform. 46% fewer manual tickets, 43% more volume handled.

Context

FinTech / banking platform with growing support volume, fragmented channels, and regulated constraints (accuracy, escalation rules, compliance). Support quality directly affected retention and customer trust.

My role

Senior PM, owner of the problem from ticket data to rollout.

Key decisions

• Fixed routing before adding AI. Automating a broken routing layer would have sped up wrong answers.

What shipped

• Rebuilt request taxonomy and ownership map. • Escalation paths for high-risk cases. • Request analytics feeding the knowledge base backlog.

Result

• Resolution time: 2.5 days → 6 hours. • Manually handled requests: −46%. • Requests handled: +43%.

Stack

AWS BedrockClaudeRAG