Applied AI Researcher

Description Join Onyx Security and help build the control plane for autonomous AI in a fast-moving, high-impact environment. Onyx is building the control layer that enables enterprises to safely adopt and scale AI agents across the organization, with the visibility, governance, and operational control needed to understand what agents exist, what they do, how they behave, and what risks they introduce.

About the role

Onyx is looking for an AI Researcher to help build the intelligence behind our AI security platform. This is a hands-on research role focused on understanding how modern large language models and AI agents behave, identifying new attack vectors and protection techniques, and translating cutting-edge research into production-ready capabilities.

The intelligence developed by this team is a core differentiator of our platform and a key driver of the company's success. You will work at the intersection of machine learning, LLMs, and cybersecurity, partnering closely with AI engineers and product teams to develop technologies that directly power our product. We're looking for someone who enjoys understanding how neural networks and foundation models work, pushes the boundaries of AI research, and is excited to see their work deployed into production and used by customers.

What you will do

Research the behavior, capabilities, and limitations of modern large language models and AI agents. Design novel techniques for detecting, evaluating, and proactively protecting against AI threats, prompt injection, model misuse, and emerging attack vectors. Develop new approaches for securing AI systems and improving the intelligence that powers our platform.

Design and run experiments to evaluate model behavior, validate new ideas, and measure the effectiveness of protection techniques. Build research prototypes and turn them into production-ready capabilities. Work closely with Product and Engineering to translate research into scalable product features. Stay current with advances in LLMs, foundation models, post-training techniques, AI security, and AI safety.

Contribute to the long-term technical direction of Onyx's AI security platform.

Requirements

What you will bring

3+ years of experience in AI research, applied machine learning, or a related field. Strong understanding of deep learning, neural networks, transformer architectures, and large language models. Hands-on experience training, fine-tuning, evaluating, or post-training modern machine learning models. Strong Python skills and experience with machine learning frameworks such as Py.

Torch. Experience designing rigorous experiments and evaluating model performance. Ability to translate research into production-ready systems. Strong analytical thinking and excellent communication skills.

Nice to Have

Experience with AI security, AI safety, adversarial machine learning, or robust machine learning. Experience developing or evaluating autonomous AI agents, reasoning models, or agentic workflows. Experience with supervised fine-tuning, model optimization, post-training techniques, or reinforcement learning. Publications in leading AI conferences (NeurIPS, ICML, ICLR, ACL, EMNLP, etc.)

or impactful applied research in industry. Experience deploying machine learning models into production and building scalable AI-powered products.