PCG logo
Case Study

Simplifying Complex Services: AI in Action at SIMT

customHeroImage
About SIMT

SIMT is a North Macedonian company delivering IT consulting and support services to both public and private sector clients. As a key local technology partner, SIMT handles a significant volume of client interactions and official documents, including customs and invoice records. With a history of reliable service delivery, the company is seen as a trusted advisor in the region. As its client base and service complexity grow, SIMT continues to seek modern, scalable solutions to maintain its high standards.

The Challenge

SIMT needed to modernise its core operations by leveraging artificial intelligence to streamline its services and enhance both internal efficiency and customer experience. With high volumes of official customs documents and frequent client queries, SIMT’s consultants faced heavy workloads and complex demands. The company aimed to reduce overhead, improve accuracy, and provide faster, more personalised responses to both customers and internal staff—all without lengthy development cycles or deep AI expertise in-house.

Business Objectives:

  • Reduce manual processing of customs and invoice documents.
  • Improve response speed and consistency for customer and consultant queries.
  • Minimise technical debt by using scalable, low-maintenance solutions.

Technical Objectives:

  • Develop three AI-powered agents to assist employees, support customers, and automate document processing.
  • Ensure secure, cloud-native implementation using Amazon Web Services (AWS).
  • Prioritise scalability, fast delivery, and integration potential.
The Solution

PCG designed and delivered a production-ready proof-of-concept (POC) built entirely on AWS, deploying three distinct serverless AI agents tailored to SIMT’s use cases. Each agent was powered by Amazon BedrockExternal Link using Claude Sonnet models and tightly integrated with other AWS services.

Key components of the solution

1. Employee-Facing AI Agent:

  • Natural language interface trained on internal support and training documentation.
  • Deployed using Amazon API GatewayExternal Link, AWS Lambda, Amazon Bedrock, Amazon OpenSearchExternal Link, and Amazon S3External Link.
  • Helps consultants quickly find answers and resolve client cases.

2. Customer-Facing Chatbot:

  • Interactive FAQ-style chatbot embedded on SIMT’s websiteExternal Link.
  • Processes natural language queries from customers, returning accurate, real-time answers.
  • Powered by scalable serverless infrastructure for high availability.

3. Document Processing Agent:

  • Analyses customs declarations, invoices, and similar documents.
  • Extracts and structures data using Claude Sonnet via Amazon Bedrock.
  • Output results in a consistent JSON format for integration into backend systems.

All agents were deployed using Infrastructure-as-Code (IaC) templates, enabling repeatability and easy handover. The solution was also designed with AWS BudgetsExternal Link in place to track cost control and maintain transparency throughout.

Additionally, as AWS Premier Tier Services Partners, we were also able to secure funding for the full POC, making this an extremely cost-effective way for SIMT to test whether this approach would fit their business needs.

Architecture diagram
Architectural diagram
Results and Benefits

The cloud-native AI agents were up and running in a short time, with little need for manual intervention. From the outset, they handled queries with clarity, integrated cleanly into SIMT’s infrastructure, and showed strong alignment with the knowledge base content. Technical features delivered as part of the solution included:

  • Fully operational AI-powered services within the AWS cloud.
  • Secure, scalable architecture supporting high request volumes with minimal operational overhead.
  • Custom fine-tuning for each model, delivering accurate, context-aware responses.

Smarter Workflows and Real Business Impact

Beyond the technical success, the implementation brought clear business value. The solution improved employee workflows, strengthened customer satisfaction, and automated previously manual and repetitive tasks.

  • Faster Case Resolution: Consultants gained instant access to structured support material, improving turnaround times.
  • Better Customer Experience: Customers received more consistent and immediate answers to their queries via the website chatbot.
  • Process Automation: Document evaluation tasks that previously required manual review were automated, freeing staff for higher-value work.

The POC proved the business case for AI-enabled consulting and operational support. With this foundation in place, SIMT can scale the system further or integrate it with additional business systems.

About PCG

Public Cloud Group (PCG) supports companies in their digital transformation through the use of public cloud solutions.

With a product portfolio designed to accompany organisations of all sizes in their cloud journey and competence that is a synonym for highly qualified staff that clients and partners like to work with, PCG is positioned as a reliable and trustworthy partner for the hyperscalers, relevant and with repeatedly validated competence and credibility.

We have the highest partnership status with the three relevant hyperscalers: Amazon Web Services (AWS), Google, and Microsoft. As experienced providers, we advise our customers independently with cloud implementation, application development, and managed services.

Author

Author Photo

Robert Spittlehouse

Content Writer
With a background in marketing and web development, Robert writes about a healthy range of cloud and digital themes, making technical detail readable. He prefers clarity, cats, and flat hierarchies—while quietly overthinking the ways technology shapes how we live.

Services Used

Continue Reading

Article
AI
How Generative AI Is Powering Real Business Impact

A comprehensive look at how Generative AI is being applied to real business challenges, with practical examples from customer engagement, data analysis, and retail optimization.

Learn more
Article
AI
From Pilot to Production: A Roadmap for GenAI on AWS

A practical, phase-based guide for turning Generative AI pilots into production-ready solutions using AWS, with a focus on strategy, governance, and scaling.

Learn more
Article
Education
Education
AI
From Storage to AI: 5 Cloud Use Cases for Academia

An informative article on how cloud computing transforms academic institutions, covering five key use cases from data management to AI-enhanced research.

Learn more
Case Study
Education
AI
Enabling AI-Driven Research at RCCL with AWS

A research lab at the University of Athens modernised its infrastructure to support AI-driven robotics research, using AWS services such as SageMaker, Bedrock, and IoT Core for scalable experimentation.

Learn more
See all

Let's work together

United Kingdom
Arrow Down