I build production-grade AI systems specializing in Conversational AI, RAG Pipelines, and Real-time ML Applications.

• 🤖 Generative AI Engineering: Building LLM-powered applications with RAG architectures
• 💬 Conversational AI: Developed voice-powered support systems achieving 92% transcription accuracy
• 🔍 Semantic Search: Engineered RAG pipelines with 95% recall for developer Q&A
• 👁️ Computer Vision: Real-time ML systems with sub-500ms latency for production deployment

AI/ML Frameworks

Generative AI & NLP

Backend & APIs

Cloud & Deployment

AI-Driven GitHub Repository Exploration Tool

• Built RAG pipeline using Gemini Pro with 95% recall for developer Q&A
• Engineered semantic search with sentence-transformers and Supabase pgvector
• Serverless React deployment for instant codebase analysis

Tech: LangChain Gemini Pro Vector Databases React Supabase

Real-time Drowsiness Detection System

• CNN architecture in PyTorch achieving 83% classification accuracy
• Implemented GANs and Diffusion Models for data augmentation
• Real-time inference with sub-500ms latency for production use

Tech: PyTorch Computer Vision CNNs Real-time ML

Voice-Powered Virtual Assistant

• Conversational NLP engine for multi-turn dialogue
• React-based responsive interface
• Local deployment with privacy-first architecture

Tech: React NLP Conversational AI Python

Generative AI Intern @ Ernst & Young (EY)

• Developed voice-powered conversational support system
• Achieved 92% transcription accuracy using OpenAI Whisper API
• Improved resource retrieval efficiency by 69% for 150+ daily queries
• Built AI automation workflows saving 10+ hours weekly

• Advanced RAG architectures and optimization techniques
• Multi-agent AI systems with LangChain
• Production MLOps and model deployment strategies

💡 Open to opportunities in AI Solutions Engineering, LLM Application Development, and Conversational AI roles.

⭐️ From AryamanGupta001