Offensive Cyber Researcher

We are looking for an experienced Offensive Cyber Researcher to reproduce and analyze publicly disclosed vulnerabilities and build controlled, reproducible evaluation environments for AI systems. The role focuses on developing offensive-cyber research tasks and scoring frameworks to assess AI models' ability to analyze and remediate vulnerabilities. The position is fully remote (based in Israel) and part of a team advancing AI cybersecurity research.

Mission

Own the CVE Exploitation Gyms and Offensive Cyber research areas. You will take disclosed, public vulnerabilities and build the reproducible lab environments and scoring harnesses that let us measure, in a controlled setting, how capable AI systems are at analysing and remediating them. All work is defensive in intent and performed exclusively against infrastructure we own.

Key responsibilities

• Reproduce disclosed CVEs end to end in isolated lab environments and capture each as a versioned, containerised evaluation task with a deterministic scoring harness.
• Design offensive-cyber evaluation tasks spanning the attack lifecycle and author the graders that score a model's performance on them.
• Write reference solutions and rubrics so that every task has a known-good outcome and a defensible score.
• Maintain strict isolation, authorisation and data-handling controls for all lab infrastructure.
• Document findings and publish internal and, where appropriate, external research on model cyber capability.

Required Skills & Experience (Must Have)

• M.A. / M.Sc. in Computer Science from an accredited university (degree completed; thesis track preferred).
• Prior hands-on experience at a cyber security company - a minimum of three years in a research, security engineering, threat-intelligence or fraud / financial-crime role at a company whose core business is cyber security.
• Demonstrated vulnerability-research or red-team experience, including reproducing CVEs from public advisories and proof-of-concept material.
• Strong systems fundamentals: Linux internals, networking, memory-safety classes of bugs, web and application security.
• Fluency with containerisation and infrastructure-as-code (Docker, and tools such as Kubernetes or Terraform) to make environments reproducible.
• Scripting and automation in Python; comfort reading C / C++ and at least one systems language.
• A rigorous, safety-first mindset regarding authorisation, scoping and responsible handling of exploit material.

Advantageous

• Published CVEs, CTF authorship, or recognised contributions to open-source security tooling.
• Experience building or operating cyber ranges, CTF platforms or automated exploitation pipelines.
• Familiarity with evaluating LLMs or autonomous agents on security tasks.

What We Offer

• A frontline role shaping how the industry measures and governs AI cyber capability.
• A senior, collaborative research team and direct access to frontier models and tooling.
• Support for responsible publication of your research

About us

With more than 2,000 customers and operations in 13 cities around the world, Innodata is the AI technology solutions provider-of-choice for 7 of the world’s biggest technology companies, as well as leading companies across financial services, insurance, law, and medicine. We’re at an exciting stage in the company and fortunate in these times to grow our team.

We’re an equal opportunity employer.

Only relevant applicants will be contacted