Automated ticket analysis with AI | Protos technology – the cloud experts

Recurring manual tasks bind time in many IT departments and qualified resources-especially in the area of ​​IT security. In this way, vulnerability scanners regularly create new entries in central ticket systems such as Jira or ServiceNow. The manual viewing and classification of these tickets are complex, error -prone and hardly scalable.

In a current customer project, we showed how generative AI (Genai) can be used beyond classic chatbots – for intelligent automation of real business processes. The focus: Safe, cloud -based and compatible with the requirements of public organizations.

RAG, Embeddings & AWS: The technological approach

The heart of the solution is a so-called Retrieval Augmented generation (RAG) approach. New content – in our case: ticket texts and content – are compared with existing, already assessed entries. The aim is to automatically identify similar tickets semantically and derive follow -up campaigns.

So -called embeddings are used: text contents are translated into mathematical vectors, which enable it to compare arithmetic similarities. Vectors can be imagined here just as in school, only that they do not include 2 or 3 dimensions, but 256 dimensions. This is particularly helpful if the same content is formulated differently.

Our solution is based entirely on services from the AWS region Frankfurt. Are used, among other things:

• Amazon Bedrock (Titan) To produce the embeddings. Titan is a Large Language Model (LLM) trained by AWS, which is relatively compact and therefore fast and cheap. It also runs completely in the AWS region of Frankfurt. Automatic drift detection ensures that code and infrastructure always remain synchronous. No manual rework required.

• Postgresql with vector expansion (on Aurora serverless V2) for semantic comparison

• AWS Lambda and Sns For the orchestration of the data flows

A special advantage of Aurora serverless V2: The database automatically scales with the load – and even reduces zero in the event of inactivity. This saves costs and makes the solution particularly economical for volatile usage patterns such as tickets, which are often collected to Jira.

Architecture diagram of automated ticket analysis

The core is a Lambda function called via Jira Automations via an SNS cue. This in turn calls Amazon Fedrock for the inference and stores the data in an Aurora serverless PostgreSQL. The database password lies in the Secrets Manager. The KMS manages a key to secure the “At Rest” files with an individual key. S3 is mainly used for the commissioning of the Lambda function and the associated layer.

Intelligent ticket conduction in practical operations

Our customer – a large global group with a high IT focus – had a clear goal: the evaluation and forwarding of security -relevant tickets must become more efficient. Together with the IT team, we have developed a AI-based system that analyzes new weakness tickets, compares with existing and automatically assigned a responsible team.

The advantages are obvious:

• Time saving: The manual review is no longer necessary for a large part of the tickets.

• Consistency: Similar tickets are treated equally – regardless of daily form or responsibility.

• Compliance: The solution meets the requirements of the customer -specific company directive for dealing with AI fully – especially since KI is not used to generate content, but only to compare the tickets.

The system runs entirely in the AWS Cloud and meets the highest data protection requirements – an important point for companies in the public sector.

Business case: added value with common sense

The technical solution is slim – the economic benefit is clearly measurable.

A short calculation example for one of the numerous IT teams in the group:

• A IT employee with € 70,000 annual salary Spend 4 hours a week with a manual ticket evaluation.
• That corresponds approx. 5,800 € per year.
• In comparison: cause the AWS services used Monthly operating costs of less than € 5 – around € 60 a year. The costs are so low because it is consistently set to serverless services (serverless services).

Conclusion: The investment in Genai is already worthwhile for smaller cases. The savings also increase scale effects.

Conclusion and outlook

Generative AI is more than a chat bot. It helps companies automatically solve real challenges – safe, scalable and understandable. At Protos technology, we rely on cloud native architectures, modular services and the highest data protection standards.

The next step: use the solution productively and roll out further processes – for example in the areas of maintenance, incident management or customer service.