Niranjan Alase

Built a Provisioning Reliability Index (PRI) engine using a weighted multi-metric algorithm with cascade deduplication to accurately attribute ESX host failures without double-penalizing dependent VMs. Designed a real-time risk classification system combining PRI scores with error severity and ESX failure thresholds to color-grade server pools across fleet, zone, and host/VM scopes.

Project Intern - Hewlett Packard Enterprise (HPE)

(2026-02 - 2026-05)

• Built a Provisioning Reliability Index (PRI) engine using a weighted multi-metric algorithm with cascade deduplication to accurately attribute ESX host failures without double-penalizing dependent VMs.
• Designed a real-time risk classification system combining PRI scores with error severity and ESX failure thresholds to color-grade server pools across fleet, zone, and host/VM scopes.

Software Development Intern - Marketrack - Remote

(2025-09 - 2026-01)

• Improved and extended the existing codebase by building scalable frontend and backend features.
• Optimized frontend API consumption for large datasets, reducing client-side processing time by 15% and improving page load performance.
• Implemented route-based lazy loading and code splitting across the dashboard and multiple frontend modules, reducing initial bundle size, improving page load performance, and delivering a faster, more responsive user experience.

Research Intern - Amity University – USRF 2025 - Noida, India

(2025-06 - 2025-07)

• Developed a Task-Specialized Knowledge Distillation (TSKD) framework for bone fracture detection by distilling knowledge from multiple anatomically specialized teacher models into a lightweight student model.
• Achieved 93.81% testing accuracy while reducing model size by 95% and GPU memory usage by 97%, enabling deployment on resource-constrained devices.

B.Tech - CSE (Major in Artificial Intelligence and Data Science) - Indian Institute of Information Technology (IIIT), Kottayam (2027)