Sr Software Engineer (Title: Performance Engineer) Resume Example

Technical Stack: Python, Spark, SQL, Git, GCP

• Worked on anomaly detection module to get huge metric data and processing for anomaly detection on runtime using python, spark and deploying on Google Cloud. Used lightGBM and STM models. In runtime, precision and recall were 85% and 99% respectively
• Worked on auto-scaling of resources by deciding how many servers need to be allocated based on past data and current trend using python, spark and deploying on GCP. Used combination of KNN and regression models and got accuracy of 90%

Technical Stack: Java, Python, C, SQL, TSDB, typescript, Angular, java APIs, CUDA, Git, GCP

• AlertCorr - Designed and developed product based on Event correlation and root cause analysis which analyses alerts coming from NetDiagnostics(APM Tool) and detects correlation and diagnoses root cause. Use combination of algorithms for frequent pattern detection, anomaly detection and learning from past data. For many cases, it analyses over 10,000 time series and provides root cause at metric level in runtime with accuracy of 88%, while for another 6% cases it comes close to actual root cause. Used java, python, SQL, jerseyAPI.
• Customer review analysis- sort bad user sessions from good ones using sentiment analysis. Designed and managed using Python
• Log analysis - find anomalous logs in real time. Using pattern matching it will find out if current set of logs which are getting generated are anomaly and pointing to potential issue. Used combination of drain tree and LCS to identify log structures which is passed to RNN. In runtime it processes 200.000 logs per minute. Designed, developed and managed using Python
• Created ML library for other products to include Machine learning in their domain. Designed and managed in java, python.
• NetDiagnostics Dashboard Functionalities - designed and developed library for applying various functionalities such as seasonal prediction, anomalies, etc. in real time. Optimized to work on large data in real time
• NetDiagnostics Alert and Anomaly Detection - singularly designed and developed optimized O(1) solution for theta prediction to get real time anomalies proactively on large time-series data

Technical Stack: Java, Objective C, Swift, C, C#, Linux, CUDA, Git, CVS, GCP

• Built iOS framework to give support of product NetVision in iOS apps, which allows app developers to monitor and replay user experience
• In-depth analysis and optimization of ELK kit to conduct faster indexing and search on big data. Added features and operations in data analysis in Java
• Developed monitors to capture system metrics in real time and also created alerting mechanism based on these metrics in Java
• Developed module which uses clustering to conduct analysis on set of flow-paths to identify patterns for performance analysis
• Developed module for parameter prediction using Machine Learning to ascertain future problem that could occur. Precision and recall for this project was 95 and 99 percent respectively