Staff ML Engineer (Zürich)

📋 Description

• • Join Lakera.Ai as a Staff ML Engineer and operate at the absolute cutting edge of Artificial Intelligence, contributing to the development of the most robust defenses in the rapidly evolving AI cybersecurity market.
• • This pivotal role involves significantly expanding the scope and enhancing the quality of our existing AI defenses, playing a crucial part in enabling the secure, large-scale deployment of Generative AI applications.
• • You will be instrumental in leading the creation of state-of-the-art machine learning models that not only perform exceptionally well but also operate with real-time efficiency, a critical factor in dynamic security environments.
• • A core aspect of your work will be the continuous identification of complex edge cases that challenge our current detection capabilities, requiring a deep dive into understanding potential vulnerabilities.
• • You will be responsible for the meticulous collection and sophisticated generation of intricate threat datasets, specifically designed to improve the performance, resilience, and accuracy of our AI defense models.
• • Your efforts will directly ensure that Lakera consistently ships the most advanced and effective defenses available in the market, maintaining a high cadence of innovation and deployment.
• • This position is for individuals whose passion lies in shaping the future of technology, focusing on groundbreaking ideas that will define the secure use of AI for decades to come.
• • Your contributions will be vital in ensuring that AI can fully unleash its immense potential for the betterment of society, by building the necessary security infrastructure.
• • About Lakera: Lakera is driven by a clear mission: to ensure AI systems behave precisely as intended. We are building the foundational security layer for a future where AI agents will manage significant aspects of our businesses and personal lives.
• • We are not merely envisioning this future; we are actively constructing the security framework that underpins it, empowering security teams and developers to confidently adopt AI technologies and usher in the next era of intelligent computing.
• • Our clientele includes Fortune 500 companies, innovative startups, and leading foundation model providers, all of whom rely on us to protect them and their end-users from adversarial misalignment and unintended AI behaviors.
• • We are also the proud creators of Gandalf, recognized globally as the most popular AI security game, demonstrating our commitment to education and engagement in the AI safety space.
• • Lakera operates with agility and intensity, fostering a culture where we act as a unified team while expecting every member to embrace substantial ownership and accountability for their work.
• • Transparency is a core value, upheld at every level of the organization, coupled with an unwavering commitment to continuously raising the bar in all our endeavors.
• • We actively promote diversity of thought, firmly believing that a wide range of perspectives is essential for achieving the best possible outcomes and driving innovation.
• • Example Projects: You might extend the capabilities of one of our core detectors to seamlessly support multiple languages without compromising performance on existing supported languages.
• • Another potential project involves increasing the scope where our defenses perform by an order of magnitude (e.g., handling multi-lingual data, code, and other complex inputs with superior efficacy).
• • You could also contribute to enhancing our modeling and data generation techniques to enable the passing of additional, critical context to our prompt injection detectors, such as LLM output characteristics and behavioral metadata.
• • Furthermore, you will have the opportunity to red team our models, proactively identifying the most pressing vulnerabilities within our detectors and prioritizing the data collection and generation efforts required to address them effectively.
• • This role offers a unique chance to influence the direction of AI security and contribute to a mission that has a profound impact on the future of technology and its responsible integration into society.