Jun 2018 – Jan 2021 · 2 yrs 7 mos · Bengaluru Area, India

• Working with infra health monitoring team.
• Worked on the whole pipeline to fetch, create, feed and show network alerts to a dashboard at scale which acts as a single pane of glass for infrastructure health monitoring. It includes creation, correlation, enrichment and auto remediation of events to provide the entire view of infrastructure and minimizing the noise.
• Worked on reasoning based correlation engine to provide a streaming, stateful correlation as a self service, providing the root cause and causal analysis of infrastructure state at scale.
• Initiated and leaded the efforts for ML experiments for easy pattern recognition and automatic clustering of network events, leading towards dynamic correlation reducing the mean time to detect(MTTD) network anomalies.
• Worked on automatic validation and remediation of network events through self service BPMN models reducing the mean time to remediate(MTTR) network anomalies.
• Worked on data platform to persist big data for longer duration which help in analyzing and ability to use visualization and ML tools over that, pushing and handling ~2M events per min.
• Technologies: Python, Java, MySQL, Redis, ElasticSearch, Kafka, ZooKeeper, Spring, Rest.li, Drools, BPMN, Jupyter Notebook.

May 2018 – Jun 2018 · 1 mo · Bengaluru Area, India

• Contributed to Red Hen Lab as a part of Google Summer of Code(GSoC) 2018, in field of computer vision using deep neural networks.
• Worked on semantic image segmentation, using Mask R-CNN on visual genome and COCO dataset.
• Worked on implementation of paper "Learning to segment every thing" by facebook using PyTorch.

Jan 2018 – Jun 2018 · 5 mos · Bengaluru Area, India

• Worked with customer communication team.
• Designed and implemented scalable system for remote repository management.
• It includes, plugin based, automated code validation system to detect potential vulnerabilities and code smells for groovy scripts.
• Implemented module for hot deployment of applications which are using those scripts, working in distributed environment.
• This solved the problem of accumulating and deploying scripts from the other teams who are using the communication system.

May 2017 – Jul 2017 · 2 mos · Bengaluru Area, India

• Worked in Production Infrastructure and Engineering(PIE) with Infrastructure health monitoring team.
• Designed and implemented scalable and flexible Alerting Engine for network monitoring.
• It was a customizable plugin based Alerting Engine with initial support of 6 plugins, including instantaneous, temporal, stateless and stateful plugins.
• The new alerting engine solved the problem of making custom plugins without touching any other parts of alerting engine, which allowed developer to create independent plugins as per the need.
• It supports per minute monitoring and alerting of more than 25M metrics data of 10K devices and 1.1M interfaces used in current LinkedIn infrastructure.

May 2016 – Jul 2016 · 2 mos · Gurgaon, India

• Worked on research and implementation of various algorithms for dimensionality reduction.
• Designed and implemented an efficient adaptive dimensionality reduction pipeline in python, to apply to various problems.
• Modified nearest neighbors algorithm for class imbalanced data.
• Gained invaluable exposure to research areas in Machine Learning.