Vanshaj Srivastava

MSc Cloud Computing graduate with practical experience in deploying and managing cloud infrastructure on AWS and Azure. Proficient in Kubernetes (EKS), Docker, IAM, and CloudWatch monitoring. Successfully designed a production-grade private cloud using AWS EKS and developed a Domain-Specific Language for multi-cloud resource automation. Strong skills in scripting, Linux, and cloud-native design patterns, with a focus on automation and security best practices.

Private Cloud Deployment using AWS EKS 

University of Leicester

• Designed and deployed a highly available private cloud architecture using Amazon EKS.
• Configured IAM roles and policies, provisioned Kubernetes clusters, and managed containerized applications via Docker and ECR.
• Set up CloudWatch for real-time logging, metrics, and alerting to improve system observability.
• Integrated S3 and ELB to ensure secure storage and load balancing.
• Achieved automated scaling and reliable microservices deployment.

AWS EKS Agile Cloud Automation

University of Leicester                                                                                                                                                

• Designed and implemented a custom Domain-Specific Language (DSL) for multi-cloud resource management across AWS, Azure, and GCP.
• Defined formal DSL grammar using Xtext, supporting resource types (VMs, Storage Buckets, Databases), configuration parameters, cost tracking, and compliance policies.
• Built an Xtend-based code generator to transform DSL models into YAML configurations for integration with Infrastructure-as-Code tools like Terraform.
• Developed an Eclipse-based editor with syntax highlighting, validation, auto-completion, and error diagnostics to improve usability for non-programmers.
• Wrote JUnit tests to validate grammar correctness and real-world cloud configuration scenarios.
• Demonstrated the tool’s capabilities through use cases in cloud inventory reporting, cost optimization, and policy enforcement for DevOps teams.

NLP-Based Interface for a Rule-Based Reasoning Engine

University of Leicester – MSc Final Project

• Designed and implemented a modular NLP pipeline to extract structured triples from unstructured text using spaCy and medSpaCy.
• Built a Graph Generation Module with NetworkX to model domain-specific knowledge in Family and Medical domains.
• Integrated Groove for rule-based reasoning over knowledge graphs with domain-specific transformation rules.
• Developed an Explanation Generation Module using OpenAI GPT-3.5/4 APIs to produce clear, human-readable justifications for inferred relationships.
• Implemented domain-specific configurations supporting modular design for new application domains.
• Conducted extensive unit, functional, and integration testing across NLP extraction, graph generation, reasoning, and explanation components.
• Demonstrated use cases in Explainable AI (XAI) for healthcare and family relationship modeling.
• Technologies: Python, spaCy, medSpaCy, NetworkX, Groove, OpenAI GPT APIs, JSON.