Exposure management and Attack Path analysis

Development of Django based services for their breach and attack simulation (BAS) platform firedrill. Integrating with other security products (SIEMs, EDRs, Vulnerability scanners, User Monitoring Tools). Performance gauging and improving the scalability of their Cloud Platform leveraging distributed architectures. Methodologies used to increase distributed processing include integrating with Kafka, celery as a task queue and Redis for caching to restrict synchronous calls in the system.

End to end development of ETL pipeline for each server installation through terraform. ETL pipeline runs in AWS Glue leveraging Spark, AWS lambda, Crawlers.

Project : Trinity

Rest API development to incorporate data from various third party vendors and provide a basis for asset selection. Deep understanding of security policies, assets, vulnerability, identity etc and providing a simple view in a single pane of glass.

* Comprehensive log analysis using machine learning to identify network health and statistics

* Auto grouping workloads and recommending policies specific to the components deployed in the network stack

* Maintaining, testing and supporting the ShieldX cloud which stores components for analysis of network and identifies threats, malwares.

* Migration of the SXCloud from AWS to GCP's Kubernetes engine.

* Neural networks designed to identify malicious URL's and patterns.

• Developed a Document Analyser using various machine learning techniques such as LDA,LSI and HDP with help of open source packages like gensim, scikit-learn and tensorflow.

• Summarization of Document (reduction in size upto one fifth of original size) with almost similar accuracy levels

• Developed a Source Code Analyser training Deep Neural Networks using tensorflow

• Identification of patterns in interactions of VMs in client network using event logs and writing ACL rules based on these patterns.

• Identify the network architecture based on the model developed from event logs