Faster Error Resolution with Automated Log Analysis
An AI pipeline that reads log files and writes error resolution summaries for a cloud networking company’s support engineers
About the customer
Our client is a US cloud networking company that simplifies and secures multi-cloud environments across providers such as AWS, Azure and GCP. When a customer’s environment has problems, its support engineers have to analyze large system and networking log files, talk to the customer, and write a summary of how the issue was fixed. This was manual and slow. We built an AI pipeline that analyses log files automatically and writes error resolution summaries to assist the engineers.
What needed solving
Vast log files
Support engineers had to analyze vast amounts of system log files during customer incidents
Manual troubleshooting
Engineers manually sifted through logs, communicated with the customer and documented resolutions
Delays for customers
The process was time-consuming and prone to delays, putting customer satisfaction at risk
How we built it
Knowledge base
AWS OpenSearch stores historical Zendesk support tickets, with their error types and resolution steps.
Issue detection
LangChain with Llama and Mistral models identifies critical issues in new log files.
Resolution summaries
Using retrieval-augmented generation (RAG), the pipeline draws on the knowledge base to write resolution summaries for engineers.
Processing pipeline
vLLM, FastAPI and Docker power the pipeline.
Deployment
Deployed in the client’s IT team infrastructure on AWS ECS, for scalability and security.
Support engineers resolve issues faster
Technologies
Domains
Have a Query? Let's Have a Conversation
Do you have a project in mind? We’d love to know what it is. Email us or send a message from our contact page.
Have a similar problem?
Tell us what you are trying to solve and our team will get back to you.