Pranali D Palshetkar
The 9th edition of the Global Testing Retreat 2024!
About Speaker
Result driven and seasoned professional with 17+ years experience in the core telecom and IT domain, specializing in telecom network infrastructure OSS, BSS,4G VoLTE architecture. Skilled in SMS center SMSC and allied Value- Added Services ( VAS) products.Roaming testing,System integration testing, API testing. Proven track record of delivery of high quality projects, vendor management,people management.
Personal Interest, Outside of work, I enjoy Reading spiritual and philosophical books Practicing yoga and meditation Listening to music and dancing too Traveling and exploring new cultures Cooking and experimenting new recipes Spending quality time with family and friends
More Speakers
Abstract in Detail:
These days Generative AI as tool is adopted & implemented for customer services and personalized customer workflows. It can respond more immaculate to customer for first contact resolution. So, with this in lieu we thought why not that we can have it in “Bug Analysis” i.e., debugging with Gen AI instead of manual debugging of logs.
Any E2E solution involves the integration between multiple components. So, while testing any E2E journey, whenever we encounter an error, as a primitive step we tend to check the “error description” based on which we can check the relevant application logs or the middleware integration layer logs connecting to the systems to figure out at which end the problem could be. This helps us to redirect the issue to the right application owner.
The idea here is to collect and maintain a repository of artifacts i.e., error codes, historic data of previously reported bugs with RCA and fix provided. This repository is then fed as an input to the GEN-AI. During the test execution phase (Automation/Manual) if any error is encountered, GEN-AI compares the error code with stored error codes and accordingly displays the root cause of the problem.
Hence with that output response, QA can directly approach the right application team to get this fixed instead of having multiple email threads to find the right application to fix the error.
Idea Outcome Result:
This introduction of GEN-AI will save at least 0.25 MD which is generally required to do the initial analysis, troubleshooting and identifying the right application.
