AI Researcher

Role overview An applied research position focused on the offensive side of artificial intelligence and machine learning security. The role centers on discovering how modern AI systems and autonomous agents can be attacked, then converting those findings into repeatable tests, detections, and tooling that other practitioners can use. It is a deeply technical research role that bridges machine learning engineering with adversarial security thinking.

Responsibilities - Investigate how large language models, agentic AI systems, and other modern ML pipelines can be exploited or manipulated in practice. - Design and prototype attacks against AI systems, including prompt-injection, adversarial input, and model-behavior abuse scenarios. - Translate research findings into reproducible tests and detections that can be operationalized across multiple engagements. - Read and synthesize academic papers, model documentation, and codebases to extract working attack techniques. - Track the AI-security field closely and publish analyses, write-ups, or threat briefs on findings that are safe to share. - Contribute to open-source tooling and community knowledge that advances adversarial ML research.

Requirements - Strong grounding in machine learning, including how contemporary models and agents are built, trained, and deployed. - A security-first mindset, with the ability to look at a system and identify how it can fail or be subverted. - Hands-on experience with adversarial machine learning or LLM security, or demonstrable depth to ramp up quickly. - Ability to read research papers or unfamiliar codebases and convert insights into working attack prototypes. - Comfort working independently on open-ended research problems with minimal supervision.

Nice to have - Published AI-security research, blog posts, or conference talks. - Open-source contributions in the adversarial ML or AI red teaming space. - Background spanning both ML engineering and offensive security disciplines.

Benefits and work setup - Remote-friendly role, with the option to work from one of the company's offices. - Independent organizational structure oriented toward the work itself rather than external reporting lines. - Deliberate, slow hiring process with an emphasis on long-term retention. - Application process asks for a portfolio (repository, write-up, tool, or track record) rather than a traditional cover letter.