Skip to content
Back to case studies
·9 weeks
Industry: fintechAI systemsCybersecurity

AI support copilot for an EU neobank — 42% cost drop, 18% CSAT lift

A fast-growing EU neobank's support team was drowning in Tier-1 tickets. We built a retrieval-augmented copilot with strict guardrails, live traffic in week four, metered against real CSAT.

THE PROBLEM

[1/3]

  • 01Tier-1 ticket volume grew 3× in 8 months; hiring couldn't keep up.
  • 02Previous vendor's chatbot hallucinated account balances — legal pulled it.
  • 03No eval pipeline; every release was a gamble.
  • 04Compliance required PII to never leave the EU VPC.

THE SOLUTION

[2/3]

  • Hybrid retrieval (pgvector + BM25 + reranker) over 12k KB articles.
  • LLM router: small model on FAQs, GPT-4o on reasoning. 3.4× cost cut.
  • Guardrails: PII scrubber, prompt-injection detector, refuse-to-answer.
  • CI eval suite with 180 golden questions; blocks deploy if factual accuracy < 95%.
  • Full observability: tokens, latency, quality, cost per tenant.

Technologies

LangGraphpgvectorOpenAIPythonPostHogOpenTelemetry

THE OUTCOME

[3/3]

  • 01Cost per resolved ticket: −42%
  • 02CSAT: +18 points on AI-handled conversations
  • 03Tier-1 deflection: 61%
  • 04Zero PII leaks during 12-week audit
  • 05Live in week 4, full rollout by week 9

Let's get started.

Send an email or book a 30-minute call.

Start a projectBack to case studies