– Developed LLM-powered evaluation engine increasing JD–resume match accuracy by 85%.

– Achieved 95% document parsing accuracy for PDF/DOCX extraction using structured processing pipelines. – Reduced AI inference latency by 70% through backend orchestration optimization. – Designed output validation framework ensuring consistent and benchmarked AI responses. – Containerized services using Docker for reproducible environments. AI-Powered Automation Platform (R&D Prototype) 2024 – Architected Python-based automation framework using Playwright for browser workflow execution. – Implemented APScheduler engine for time-based task orchestration. – Integrated Google Gemini API for context-aware AI response generation using structured prompts. – Designed layered system separating automation, AI processing, and scheduling modules. StudentVerify – Secure Multi-Layer Verification System 2025 – Engineered secure backend supporting dynamic QR validation and GPS-based proximity checks. – Implemented RESTful APIs with JWT-based role authentication. – Delivered internal production system and clean architectural MVP demonstrating system design.