Ruhi Sharma

LLM Security: Designed and developed a Python-based security gateway to detect and mitigate prompt injection attacks targeting Large Language Models (LLMs). Threat Detection: Implemented NLP-driven prompt sanitization, adversarial prompt classification, and response filtering mechanisms to improve AI system security.

Backend Engineering: Built modular backend components using FastAPI and REST APIs to support secure AI inference workflows and extensible security integrations.

Developer Tooling: Developed a Python-based cybersecurity toolkit featuring password strength analysis, cryptographic utilities, and interactive security simulators. Backend Development: Implemented backend security workflows and reusable Flask modules to automate common security tasks and improve operational efficiency.

Secure Engineering: Applied secure software development principles while designing modular utilities for cybersecurity education and security awareness.

Machine Learning: Developed supervised machine learning models for phishing detection, fake review classification, and AI-generated spam detection with up to 92% accuracy. NLP Pipeline: Performed text preprocessing, feature engineering, and behavioral profiling across more than 1,000 structured and unstructured text samples.

Security Automation: Built automated trust-scoring workflows that generated real-time security risk assessments for suspicious online activity.

Collaborated on AI, software engineering, and hackathon initiatives with student developers through workshops and technical events involving 300+ participants. Contributed to collaborative learning activities focused on emerging technologies and software development practices.