Job Responsibilities
- Collaborate with internal stakeholders to identify business needs and develop NLP/ML solutions that address client needs and drive transformation.
- Apply large language models (LLMs), machine learning (ML) techniques, and statistical analysis to enhance informed decision-making and improve workflow efficiency, which can be utilized across investment functions, client services, and operational process.
- Collect and curate datasets for model training and evaluation.
- Perform experiments using different model architectures and hyperparameters, determine appropriate objective functions and evaluation metrics, and run statistical analysis of results.
- Monitor and improve model performance through feedback and active learning.
- Collaborate with technology teams to deploy and scale the developed models in production.
- Deliver written, visual, and oral presentation of modeling results to business and technical stakeholders.
- Stay up-to-date with the latest research in LLM, ML and data science. Identify and leverage emerging techniques to drive ongoing enhancement.
Required qualifications, capabilities, and skills
- Advanced degree (MS or PhD) in a quantitative or technical discipline or significant practical experience in industry.
- Commercial experience in applying NLP, LLM and ML techniques in solving high-impact business problems, such as semantic search, information extraction, question answering, summarization, personalization, classification or forecasting.
- Advanced python programming skills with experience writing production quality code
- Good understanding of the foundational principles and practical implementations of ML algorithms such as clustering, decision trees, gradient descent etc.
- Hands-on experience with deep learning toolkits such as PyTorch, Transformers, HuggingFace.
- Strong knowledge of language models, prompt engineering, model finetuning, and domain adaptation.
- Familiarity with latest development in deep learning frameworks.
- Ability to communicate complex concepts and results to both technical and business audiences.
Preferred qualifications, capabilities, and skills
- Prior experience in an Asset Management line of business
- Exposure to distributed model training, and deployment
- Familiarity with techniques for model explainability and self-validation