Jasmine Ansari

Background

About Jasmine Ansari

Result-driven Data Scientist with expertise in data analysis, machine learning, and generative AI. At Visionwaves, fine-tuned a LLaMA 3.1 8B model for SQL query prediction, implemented continual pretraining for schema-specific data, and developed a real-time AI avatar with live streaming and dynamic interactions along with fine-tuned model with picking out the desired words after diving them into categories for better retrieval, another fine-tuned model for function calling for date and time problem and now working on fine-tuning the gaudrail problem. At WNS Global Services, built a FAISS-powered system for employee feedback analysis. Leveraged Python for survey insights at Champaran Foundation and honed data manipulation skills at MTAB Technology. Holds a B.Tech (9.5 CGPA) and an AI certification from IIT Roorkee. Proficient in Python, TensorFlow, Hugging Face, and Azure, with experience in RAG systems, encoder-decoder models, and BERT-based QnA frameworks. Passionate about AI-driven insights and process optimization.