Yash Shah

I’m a final-year B.Tech CSE (AI & ML) student with hands-on experience delivering production-ready machine learning systems in agile teams. I’ve built and deployed models across forecasting and NLP-style generation, using Python and deep learning stacks like PyTorch and TensorFlow, along with Hugging Face Transformers.

In my Machine Learning Intern role, I developed an LSTM-based stock forecasting model with 86.5% directional accuracy, optimized CNN inference latency by 20% using quantization and pruning, and served real-time predictions through a FastAPI REST API with sub-200ms response time. I also worked on end-to-end Python & SQL ETL and created Power BI dashboards that improved stakeholder decision-making speed by 30%, and I’m especially excited to keep building GenAI workflows like LLM fine-tuning and RAG pipelines with strong MLOps practices (MLflow, Docker, CI/CD) on AWS/GCP.