"I plant seeds of automation and watch them scale into cloud forests."
I am Shiwani Yaduka, a 3rd-year Computer Science student at Lovely Professional University (CGPA: 8.61), based in Bhilwara, Rajasthan.
My journey started with curiosity about how systems communicate — and led me to a deep passion for cloud infrastructure, DevOps pipelines, and agentic AI automation. I believe that great software isn't just functional; it's resilient, scalable, and beautifully orchestrated.
Having completed intensive training in Agentic AI Platform Development at LinuxWorld Informatics, I'm now focused on building real-world projects that apply everything I've learned — from cloud-native architectures to AI-powered automation. When I'm not deploying containers, I'm probably competing — and winning.
I am Shiwani Yaduka — System Architect, Cloud Engineer, and Infrastructure Strategist operating at the intersection of automation and high-availability design.
My architecture philosophy: every system should survive chaos. I design fault-tolerant, auto-scaling cloud ecosystems using AWS, Kubernetes, and Jenkins-driven CI/CD pipelines. Every microservice I ship is containerized, monitored, and production-hardened.
With Agentic AI Platform Development training complete, I'm now channelling that expertise into building production-grade projects — from AI-driven automation to self-orchestrating cloud systems. When I'm not deploying containers, I'm probably competing — and winning.
A centralized DevOps dashboard to manage remote Linux servers and multi-cloud infrastructure. Integrates ML processing, Docker/Kubernetes automation, SSH utilities, and real-time Twilio/WhatsApp messaging for non-technical users.
Automated transition from local Flask development to production-ready Docker containers with integrated pytest suites. Implements event-driven triggers and container orchestration for seamless image updates.
A cloud-native AI assistant that monitors reading inactivity via CloudWatch and sends personalized SNS email reminders. Implements predictive completion algorithms and detailed monthly reading reports.
Role-based access control using class inheritance for Admin vs Staff privilege separation. Uses STL containers (vector, map, stack) for fast lookups and undo functionality, with JSON-based persistence and MongoDB integration.
An AI-powered chatbot trained on real WhatsApp conversations using Gemini Flash. Replicates tone, emotional style, and personality for deeply personal and emotionally intelligent conversations.
An AI-powered confidence coach that analyzes live 15-second speaking sessions using webcam, hand gesture detection, smile tracking, and voice pitch analysis to generate a real-time confidence score and motivational feedback.
Won first place for collaborative problem-solving and creative thinking.
Won two summer internship project competitions.