Bakary Gibba
Back to selected work

Case study · 03

Closer AI

A production conversational sales assistant integrating backend services, AI workflows, and user-facing review.

Explore on GitHubRead documentation
Category
Conversational sales assistant
Availability
Production system · Public case study
Closer AI product interface and system workflow

Problem + approach

Overview

Sales workflows require responsive, contextual assistance that can integrate with operational systems rather than operate as an isolated chat demo.

Technology stack

  • TypeScript
  • NestJS
  • PostgreSQL
  • OpenAI

Conceptual overview

System architecture

  1. 01Sales copilot client
  2. 02NestJS API
  3. 03Conversation + workflow services
  4. 04OpenAI
  5. 05PostgreSQL
  6. 06Review dashboard

This project-specific overview is based on the published case-study architecture. Internal implementation, credentials, and deployment details remain undisclosed.

Decisions + constraints

Engineering notes

01

Engineering decisions

  • Conversation logic is mediated by backend services rather than exposed directly to the client.
  • Persistent data supports continuity across customer interactions.
  • Generated suggestions pass through user-facing review before they reach customers.
02

Technical challenges

  • Maintaining useful conversational context
  • Balancing automation with predictable system behavior
03

Results & impact

  • Customer-facing conversational workflows delivered at Apping Technology
  • Client-specific metrics are confidential
04

Lessons learned

  • Conversational products need reliable backend contracts and carefully bounded AI behavior.

Screenshots + diagrams

Project media

Closer copilot live coaching interface
Live copilot — contextual coaching during a sales conversation.
Closer team dashboard overview
Team dashboard — operational visibility across customers and sessions.
Closer document knowledge-base screen
Knowledge base — approved documents ground product recommendations.

Availability + roadmap

Documentation

Repository & API documentation

View repositoryRead full documentation

Future improvements

  • Conversation quality evaluation
  • Operational analytics
Next case studyOutreach Sentinel