Skip to content
AI2025

Signal Support Copilot

A retrieval-grounded assistant that drafts support replies from a company's own knowledge.

The Challenge

Support volume was growing faster than the team could hire. Off-the-shelf chatbots either hallucinated or deflected everything to a generic FAQ, and agents didn't trust them.

The Solution

We built a retrieval pipeline over the company's own content with strict grounding: the assistant answers only from retrieved passages, always cites them, and routes low-confidence questions to a human agent together with the supporting material it has already gathered.

Draft with citations
Escalation handoff view
Key Features
01

Grounded drafting

Replies are generated only from retrieved, cited source passages.

02

Confidence gating

Low-confidence questions are escalated with context instead of guessed at.

03

Answer review loop

Agent edits feed back into evaluation so quality is measured, not assumed.

04

Knowledge gaps report

Surfaces the questions the knowledge base can't yet answer.

Architecture
  1. 01Frontend

    Agent console embedded in the existing helpdesk via extension

  2. 02API Layer

    Streaming completion API with per-tenant rate limits and PII redaction

  3. 03Services

    Ingestion, embedding, retrieval and a LangGraph orchestration layer

  4. 04Database

    Postgres with pgvector for embeddings; document store for source content

  5. 05External Integrations

    Helpdesk, knowledge base and model providers

Technology

Frontend

  • Next.js
  • TypeScript
  • Tailwind CSS

Backend

  • Python
  • FastAPI
  • LangGraph

Database

  • PostgreSQL
  • pgvector

Cloud

  • Azure Container Apps
  • Blob Storage

AI

  • Claude
  • OpenAI embeddings
  • Ragas evals

DevOps

  • Bicep
  • GitHub Actions
  • Langfuse
Results
  • Every answer traceable to a source
  • Low-confidence questions escalated rather than answered speculatively
  • Knowledge gaps made visible to the content team
  • Evaluation built in from the outset
Gallery
Draft with citations
Escalation handoff view
Knowledge gap report
Next project

FieldKit Mobile

View Case Study