Home > Exploring Amazon Bedrock to transform ITSM customer experiences in SolarWinds Service Desk

Exploring Amazon Bedrock to transform ITSM customer experiences in SolarWinds Service Desk

Practical IT Service Management (ITSM) can be challenging when considering the growth of users and services to support. Additionally, ongoing trends in distributed enterprises with remote workforces, hybrid cloud, and cost-based pressures force companies to do more with less. Atop the usual demands, ITSM teams increasingly expect integrations with observability and security, giving IT support teams a single pane of glass across their roles and responsibilities.

SolarWinds Service Desk is an ITSM solution built to maximize service desk staff productivity, automate service delivery, streamline ticket communications, and accelerate ticket resolution. SolarWinds Service Desk’s comprehensive features include management of tickets, assets, licenses, automation, change control, and an employee service portal and knowledge base. SolarWinds Service Desk uses Artificial Intelligence (AI) today to drive a better user experience and more efficiency across service desk teams. From our Smart Suggestions and Virtual Agent helping end-users get to resolution quicker to our Category Recommendations, Ticket Linking, and Sentiment Analysis to help agents resolve open tickets faster, we’ve embedded AI across the platform. SolarWinds is now exploring using Amazon Bedrock to enhance our AI further with new generative AI capabilities.

Driving a better customer experience

We collaborated with our customers to identify specific challenges and opportunities for improving agent experiences. Customers shared that they spent considerable time on repetitive tasks such as responding to similar tickets, creating knowledgebase articles, or summarizing incidents for retrospectives. Their teams devote significant time to writing, taking away time to resolve other incidents and respond to new requests. While the outcomes of these actions are helpful, the time spent is incredibly consuming.

Because ITSM teams are already resource-constrained and strapped for time, we started exploring how we could help these teams a few months ago. We realized generative AI would be the best answer; it was now a matter of “how” to do it. We explored various vendors and considered building it all in-house until we saw what Amazon Bedrock had to offer.

Amazon Bedrock

We evaluated different generative AI approaches to summarize ticket resolution actions and generate ticket responses and decided upon Amazon Bedrock for several reasons. Amazon Bedrock is a fully managed service that offers choice of high-performing foundation models (FMs) from leading companies via a single API, along with a broad set of capabilities you need to build generative AI applications, simplifying development while maintaining privacy and security. Amazon Bedrock’s model choices from AI21, Anthropic, Meta, Amazon, and Stability AI enabled us to satisfy complex dynamic ITSM use cases. Additionally, its system is secure, intuitive, fast, reliable, and cost-effective. Bedrock simplifies our infrastructure and allows us to focus on building and scaling our Generative AI use cases. Bedrock also allowed us to natively integrate with our existing AWS-based services in SolarWinds Service Desk and build a minimum viable product (MVP) made available to customer focus groups for immediate product feedback and early access.

SolarWinds AIOps - Generative AI Architecture

Our Secure by Design initiative, incorporated into our technological decisions and processes, guided our direction. In addition to our exploration of Amazon Bedrock, we’ve developed an LLM Gateway Service, ensuring user and customer data that leverages any foundational model is anonymized or pseudonymized and has appropriate logs before accessing a foundational model. This LLM Gateway Service safeguards sensitive information, such as Personally Identifiable Information (PII) or confidential data, before it reaches the models. Any customer content processed by Amazon Bedrock is encrypted and stored at rest in the AWS region where Bedrock is used, maintaining our Data Residency commitment to customers where applicable. Neither user inputs nor model outputs are shared with third-party model providers. Amazon Bedrock integrates with multiple AWS security services to help with our Secure by Design and compliance requirements.

What we are working on

Generative AI will be a prominent feature on our roadmap in 2024, but we must do so securely and responsibly in a way that truly benefits our customers. We’re excited to be enhancing our AI capabilities across our platform. The first few areas we’re tackling are related to generating suggested responses, case summarization, and knowledgebase article creation. These areas continue to take up valuable time from busy agents better used to resolve important incidents and respond to requests. We’ll continue to expand on these capabilities and share more of our strategy and plans in this area.

You can review the SolarWinds reports on AIOps and ITSM transformation or sign up for a Service Desk trial.

SolarWinds is an AWS Partner Network member with DevOps competency and a leading provider of simple, powerful, secure Observability and IT management solutions. You can learn more about SolarWinds in AWS Marketplace.

Sean Sebring
Some people call him Mr. ITIL - actually, nobody calls him that - But everyone who works with Sean knows how crazy he is about…
Read more