All work
AIDeveloper Tools2025

Lumen — AI Support Copilot

A RAG-powered support copilot grounded in 4,000 pages of docs — with eval suites proving accuracy before launch.

-63%
Ticket volume
94%
Answer accuracy (evals)
11s
Median resolution
Lumen — AI Support Copilot — hero

The challenge

Lumen's support team drowned in repeat questions already answered in their docs. Off-the-shelf chatbots hallucinated API parameters — worse than no bot at all.

The solution

Custom retrieval pipeline over versioned docs with cited answers, confidence-gated escalation to humans and a 600-case eval suite run on every prompt change. Deflection hit 63% without a single hallucinated API in production.

Stack

  • Claude API
  • RAG
  • Next.js
  • pgvector
  • LangSmith
Lumen — AI Support Copilot — detail 1
Lumen — AI Support Copilot — detail 2
The eval-first approach sold us. We can prove the bot's accuracy to our own customers — that's a feature, not just support tooling.
Priya RamanVP Product, Lumen Software

Start a project

Tell us what you're building

Share your brief and get a scoped, fixed-price proposal within 48 hours — including timeline, team and deliverables. Free, no obligation.

  • Fixed-price scope in 48h
  • Senior-only project team
  • Weekly shipped increments

Prefer chat? WhatsApp us at +60124760876

Scoped proposal within 48 hours. No spam, no obligation.