Amaresh Hebbar

I’m an AI/ML engineer with production experience architecting and shipping LLM-powered products end-to-end—model fine-tuning, backend infrastructure, and mobile integration. I led engineering delivery for a health-tech platform spanning document extraction, ICD-10/insurance intelligence, and 3D organ visualization, and I mentored 10 interns across the AI product suite.

I’ve published 20+ fine-tuned models on HuggingFace across medical, retrieval, and code domains, and I thrive in end-to-end delivery loops: from dataset curation and RLHF-style workflows to deployment and evaluation. At Studio Ilios/Humanova Minds, I shipped OceanAI with validation from 47 external testers, built on-device local LLM inference with cloud fallback routing, and maintained ~80% fully-offline sessions at a 45ms baseline latency with monitored degraded fallback.