nomaan bhimani

I’m an AI/ML Engineer specializing in computer vision and deep learning, with a strong bias toward practical, deployable systems. I enjoy designing architectures like CNNs and Vision Transformers, turning model behavior into reliable, confidence-governed predictions that can operate under real constraints.

My work includes a patent application for “Handheld, Adaptive and Confidence Governed Skin Disorder Analysis,” where I designed a dual-path ViT with a quality-gated, confidence-regulated classification pipeline. I also deployed the full inference stack on Raspberry Pi 4B with adaptive hardware-level image optimization, and implemented temporal feature-level memory for longitudinal evaluation. I build systems that produce not only predictions, but usable evidence and downstream workflows.

During my Data Science internship at Netleap IT, I built end-to-end Python ML pipelines using Scikit-learn, NumPy, and Pandas, and developed production-ready visualizations and dashboards for stakeholders. Across my projects, I’ve shipped ONNX edge deployments, forensic reporting pipelines with ReportLab, and RAG-based legal and medical assistants using hybrid models—while continuously refining performance through evaluation, hyperparameter search, and feedback loops.