At HNS Technologies, I build enterprise AI systems across LLMs, NLP, computer vision, and speech. I productionized a Whisper-based speech-to-text engine on AWS, reducing latency and improving operational efficiency.
I also developed driver drowsiness detection and FaceNet-based recognition systems, along with RoBERTa emotion analysis and BERT summarization models. My work includes multimodal pipelines and AI microservices deployed with FastAPI, Docker, and CI/CD.
For customer insights, I developed a Voice of Customer dashboard using Streamlit and FAISS to support semantic search across customer records. I also built an edge-optimized speech-to-speech system for offline, real-time processing on Jetson Nano.
Previously, at Alphind Software Solutions, I developed AI solutions for healthcare and enterprise clients, including a healthcare chatbot, IoT anomaly detection, and transformer-based sentiment analysis. My projects include AI-powered service recommendations, review handling, and vehicle configuration tools.
