mahesh User

I’m an AI/ML Engineer and published researcher specializing in end-to-end machine learning systems, NLP, and deep learning architectures. I focus on turning model ideas into reliable pipelines and production-ready components.

I’ve contributed to production ML repositories as a GirlScript Summer of Code 2026 (GSSoC'26) contributor, with merged pull requests and a GitHub Pull Shark achievement. I’m also the first author of a peer-reviewed paper on Emotionally Aware AI and the creator of myMLlib, an open-source ML library.

My work spans NLP, computer vision, and classification systems, with demonstrated 92-99% model accuracy across multiple domains. In projects like an end-to-end customer intelligence platform (FastAPI endpoints + clustering intelligence) and an emotional-support GPT-style model trained on 52,137 pairs, I prioritize measurable performance and practical deployment.

I’m proficient across Python and major ML frameworks (PyTorch, TensorFlow, scikit-learn) and I enjoy building strong data-to-model-to-API flows. I also care about ethical AI design—human-AI interaction, emotion safety, transparency, and bias mitigation are part of how I engineer solutions.