Applied Science - Cybersecurity

Description

The company provides full-stack AI solutions, from frontier models and developer tools to applications and compute.

It partners with enterprises in finance, manufacturing, defense, healthcare and the public sector, building customized AI systems they can run on their own terms.

The team is distributed across Europe, North America, Asia and the Middle East, and is made up of researchers, engineers, product builders and former founders.

The company is hiring Applied Scientists to build frontier cybersecurity models that cover vulnerability analysis through incident response and remediation.

You will build the environments, training data pipelines and evaluations that make these capabilities concrete and measurable, working with real production data and cybersecurity practitioners.

You can come from security with an interest in AI science, or from AI science with an interest in security. Curiosity about the side you know less, and the drive to go deep on it, matter most.

Responsibilities

• Develop, pre-train and post-train foundation models specialized for agentic defensive security.
• Tackle a range of hard, real-world problems in order to shape AI for cybersecurity.
• Create evaluations and benchmarks that test security capability on tasks representative of customer work.
• Build sandboxed execution environments, RL tasks and verifiers based on real customer workflows, and produce and curate training data for frontier models.
• Build the tools and frameworks needed for data generation, model training, evaluation and deployment.
• Own technical delivery across project sprints and handle research communication with customer teams.

Requirements

• Speak and write fluent English, and explain complex technical topics clearly to both technical and non-technical people.
• Have built harnesses, benchmarks or RL environments for code, agentic or security tasks.
• Have hands-on security experience in at least one of: vulnerability research, exploit development, red teaming, or detection and incident response.
• Are comfortable working in a large codebase and can find your way around it with little guidance.
• Write clean, readable, high-performance and fault-tolerant Python.
• Move forward without a roadmap and act with confidence when things are unclear.
• Are low-ego, collaborative and keen to learn.
• Have a record of results from personal projects, professional work or academia.
• Hold a PhD or Master's in a relevant field such as Mathematics, Security or Machine Learning. Exceptional candidates from other backgrounds are welcome.
• Have contributed to a large open-source or industry codebase.
• Have published in top ML or security venues, or contributed to open-source security or observability tools.
• Have worked on large codebases or large machine learning models at scale.
• Enjoy improving existing code through typing, tests and CI pipelines.

Conditions

Benefits differ by country. Check the benefits page for your location for current details.

Benefits

• Benefits vary by country and may include healthcare coverage, parental leave, retirement plans, relocation support, wellness programs, meal and transportation allowances, and other location-specific perks.