Staff LLM Data Scientist, TIFIN

Responsibilities:

• Design and fine-tune open source and proprietary LLMs for various tasks such as answering questions, summarization, reasoning and planning, etc.
• Build advanced Retrieval Augmented Generation (RAG) pipeline including rewriting, embedding fine-tuning, hybrid search, reranking, knowledge graphs, etc.
• Implement a comprehensive evaluation framework and metrics for model performance
• Deploy models into production environments and ensure low latency, reliability, and scalability.
• Collaborate with product team and software engineering team to build end-to-end product systems.

Requirements & Skills:

• Ph.D./Masters/Bachelor’s degree in computer science, mathematics, statistics, engineering, or relevant field
• Experienced in the field of NLP/LLM and well-versed with the current and latest state-of-the-art research
• Hands-on experience in various LLM fine-tuning techniques (e.g. LORA), LLM inference frameworks (e.g. vLLM), advanced RAG pipelines
• Excellent knowledge of LLM evaluation methods and metrics
• 5-6 years of machine learning/deep learning experience within frameworks such as TensorFlow and/or PyTorch
• 2+ years of practical experience in the development of generative AI applications
• Publications at reputable machine learning conference or journal
• Proficient in Python and SQL
• Analytical and problem-solving skills
• Ability to visualize data in the most effective way possible for a given project or study
• Thrives in a highly demanding, entrepreneurial, and fast-paced environment
• Is a top performer and has a proactive, “doer”, and problem-solver mentality
• Is highly flexible, has a good tolerance for ambiguity, and can quickly adapt to changing priorities
• Is an exceptional team player with solid communication skills