Who We Are
The Cybersecurity Group at the Qatar Computing Research Institute (QCRI), part of Hamad Bin Khalifa University and the Qatar Foundation, is in Education City, Doha, a hub for research, innovation, and higher education.
We conduct research in AI safety and security, AI for security, and cyber physical system security, addressing both foundational challenges and practical applications. We collaborate with international institutions, industry, and government to translate research into deployable solutions.
QCRI Projects:
- Fanar: Arabic-focused AI and large language models.
- aiXamine: An evaluation platform for LLM safety and security.
- AIDR: Humanitarian crisis management system used by the UN.
- ATS: Speech recognition and translation systems.
About the Research Project
This project spans two connected directions: AI safety, security, and privacy, and the application of AI to security problems.
The first direction addresses privacy and security challenges in large language models (LLMs), particularly in applications using agents and retrieval-augmented generation (RAG). This includes memorization and leakage of sensitive data, model inversion and extraction attacks, exposure of proprietary information through RAG and agent workflows, prompt injection and jailbreak attacks, reliable agent-based decision-making, and the integrity of RAG pipelines.
The second direction applies AI to security problems such as vulnerability research, code analysis, and program verification, using LLMs and agents to reason about real-world codebases and detect, analyze, and remediate security flaws.
As a Research Engineer, you will build prototypes and tools that support both directions and integrate them into aiXamine and other frameworks developed by the group. The role emphasizes engineering and implementation alongside research contributions.
Your Role
- Design and implement prototypes and tools using appropriate methods, frameworks, and techniques to meet project goals.
- Build and maintain evaluation pipelines, benchmarks, and infrastructure for LLM safety, security, and privacy.
- Develop tooling that applies LLMs and agents to security tasks such as vulnerability detection, code analysis, and verification.
- Collect and analyze research data to support experiments and findings.
- Contribute to reports, documentation, and presentations related to research activities.
- Collaborate with researchers to move results from prototype to deployed tooling.
Minimum Qualifications
- MSc in Computer Science, Cybersecurity, or a related field.
- 1-2 years of experience in software engineering or applied research.
- Experience in AI safety, AI security, or AI privacy.
- Experience in security (software or systems, not hardware).
- Strong programming skills in Python and at least one of Java, C, or C++.
- Experience building and maintaining software systems or research prototypes.
Preferred Qualifications
- Experience in software security, code analysis, or program verification.
- Experience training or fine-tuning LLMs.
- One or more publications at peer-reviewed top venues.
- Experience with agent frameworks, RAG systems, or LLM evaluation pipelines.
- Experience applying AI to security problems such as vulnerability research.
- Experience contributing to open-source software.
- Strong communication and collaboration skills.
What We Offer
- Competitive international salary with housing and transport allowances.
- Medical coverage.
- Permanent contract.
- Computing infrastructure including GPUs and private cloud.
- Collaborative research environment.
How to Apply
Send your CV and a one-page cover letter highlighting your achievements to cs-jobs@qcri.org with [Research Support Engineer] in the email title/Subject.
Application deadline: open until filled.