Virtue AI is looking to build the essential AI security platform with advanced and comprehensive guardrails and red-teaming tools to enable organizations to deploy AI applications confidently and responsibly. The company aims to define and lead the new category of AI Security.
Requirements
- Proficiency in programming languages such as Python, along with expertise in LLM libraries like PyTorch, DeepSpeed, OpenRLHF, vLLM, verl, etc
- Experience in LLM finetuning for different modalities
- Experience in building LLM-based agents for various applications
- Experience in conducting large-scale red-teaming for LLMs and agents
- Hands-on experience with Docker for containerization and deployment
- Hands-on experience in back-end engineering (Go, C/C++) and front-end development (Node.js, Typescript)
Responsibilities
- Develop our core techniques for agent and model red-teaming, including discovering novel testing paths, designing novel testing techniques, and developing automatic testing platforms
- Develop and train our core guardrail models for different input modalities and defense targets and goals
- Apply efficiency inference methods to reduce model latency
- Work with the research engineers to integrate our developed techniques into products
Other
- Thrive in this role if you’re excited by new technology, love solving customer problems, and can comfortably bridge the business and technical worlds.
- Enthusiasm for thriving in a fast-paced startup environment
- Strong Problem-solving Skills And Effective Communication Abilities
- Competitive Compensation Commensurate With Skills And Experience.
- Equal Opportunity Employer.
