Enosis Outsourcing logo
EndorsedendorsedCompleted

AI-Powered Customer Support Agents for Cato Networks

Services Covered:

AI and Machine Learning
Project hero — replace with case study imagery
Industry
Industry

Telecommunications

Duration
Duration

16 months

Budget
Budget

50K - 199K

Client Size
Start Date

1 January 2025

Team Size
Team Size

1-5 Employees

Engagement Model
Engagement Model

Nearshore

Project Capability Score (PCS)

gauge meter
85/100
Strong

Project Capability Score (PCS) estimates how capable deepsense.ai is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

Cato Networks

map Israel

Project Summary

deepsense.ai developed and deployed six AI-powered agents using LangGraph orchestration and AWS Bedrock (Claude 3) to transform customer support for a networking company. The system automated workflows across internal systems, extracted actions from internal data sources, and reduced time spent on tickets by around 20%. The solution achieved over 80% user satisfaction and automated connectivity-related tickets while being self-improving.

Key Challenges

  • Navigate multiple internal systems to investigate incoming Zendesk tickets and automate responses
  • Knowledge exchange and teaching the client about AI agents
  • Reduce Tier-1 ticket response time by approximately 20% through automation

Project Deliverables

  • Six specialized AI agents
  • Atomic agents for network analysis, events analysis, and configuration
  • Planner agent to orchestrate atomic agents during ticket resolution
  • Responder agent to handle conversation history and provide responses

Project Solution

deepsense.ai developed and deployed six specialized agents orchestrated via LangGraph and powered by AWS Bedrock Claude 3 models. The scope included atomic agents for specific actions (network analysis, events analysis, configuration), a planner agent to orchestrate workflows, and a responder agent that handles full conversation history to provide responses. One deepsense.ai team member was assigned to the engagement.

Project Outcome

  • Responder user satisfaction >80%
  • Actions extracted from all internal data sources (scalability)
  • Time spent on tickets reduced by 20%
  • Automation of connectivity-related tickets
  • Self-improving system

Platforms

  • CloudCloud

Tech Stack

  • Amazon Web Services (AWS)Amazon Web Services (AWS)

Client Endorsement

Overall Review Rating

5star

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“deepsense.ai has achieved over 80% user satisfaction and reduced time spent on tickets by 20%. The team has extracted actions from all internal data sources and automated connectivity-related tickets. Moreover, the system is self-improving. deepsense.ai has delivered on time and communicated well.”

Daniela Rosenstein

AI & Automation Lead

Note: This endorsement is based on publicly available client feedback from external review sources.