The position is for an ML Research Scientist specializing in Differential Privacy, working remotely for a software company that deploys large language models (LLMs) for enterprises.
The company has recently raised $15.1 million in a Series A funding round, bringing the total raised to $19.3 million.
Responsibilities include leading and managing an ML privacy vertical focused on specific domains and optimization strategies such as differential privacy and privacy-enhancing algorithms.
The role requires close collaboration with the engineering team to implement practical applications of developed algorithms tailored to client needs.
The candidate will contribute to the creation of research papers, patents, and presentations by integrating their work with contributions from other team members.
Requirements:
Candidates must have profound expertise in privacy-preserving ML methodologies.
Hands-on experience in training Differentially Private models using frameworks like Opacus and OpenDP is required.
Extensive practical knowledge in deploying various Large Language Model (LLM) models and architectures in real-world scenarios is essential, along with proficiency in leading end-to-end projects.
While extensive experience in federated learning (FL) is not mandatory, it is preferred.
Domain expertise in understanding attacks against ML models, particularly in Computer Vision (CV) or Large Language Models (LLMs), is desirable.
Adaptability and flexibility are crucial, as the role may require swift learning and implementation of new findings in the field.
Benefits:
The position offers the opportunity to work in a cutting-edge field focused on privacy and machine learning.
Employees will be part of a dynamic team that is at the forefront of research and development in ML privacy.
The role provides a chance to contribute to significant research outputs, including papers and patents.
Working remotely allows for flexibility in work arrangements.
The company is backed by reputable investment firms, indicating strong financial support and growth potential.
