Lalit Singh Kharayat

I’m a Software Engineer with hands-on experience across AI/ML, Generative AI, Retrieval-Augmented Generation (RAG), backend systems, and intelligent automation. I’ve interned at National Instruments (NI) and the Indian Institute of Science (IISc), and I focus on building production-grade pipelines and systems that are reliable under real constraints.

At IISc (SPIRE Lab), I designed and deployed scalable annotation and data management workflows for 2,000+ real-time MRI (rtMRI) and audio samples, improving data retrieval speed by 40% through optimized indexing and automated workflows. I worked on temporal alignment using Dynamic Time Warping (DTW), implemented FCN-based semantic segmentation for vocal tract boundary delineation, and engineered end-to-end ML pipelines for acoustic-to-articulatory inversion with transformer-based sequence modeling.

At NI, I engineered automated PMIC validation systems using custom pin-centric architectures and LabVIEW frameworks, reducing manual testing and improving test automation efficiency. I also optimized ADC/DAC signal accuracy and designed/deployed “Zero-Volt” e-Load systems for hardware validation in extreme edge-case conditions—while keeping a strong bias toward measurable outcomes. Outside internships, I’ve built an LLM-powered RAG document Q&A app deployed on Hugging Face Spaces, and delivered IoT-enabled edge ML projects for virtual fencing and real-time skin abnormality detection.