At a Glance
- Tasks: Join a cutting-edge research team to develop security theories for AI systems.
- Company: Oxford Witt Lab for Trust in AI, part of the Department of Engineering Science.
- Benefits: Competitive salary, opportunity for extension, and a collaborative research environment.
- Other info: Engage in innovative research with excellent career growth opportunities.
- Why this job: Make a real impact in AI security while advancing your academic career.
- Qualifications: PhD/DPhil in relevant fields with strong mathematical and programming skills.
The predicted salary is between 33900 - 39900 £ per year.
We are seeking a full-time Postdoctoral Research Assistant to join the Oxford Witt Lab for Trust in AI (OWL) at the Department of Engineering Science (Central Oxford). The post is funded by Schmidt Sciences (AI2050 Early Career Fellowship) and is fixed-term for 24 months, with a possible extension beyond this date subject to satisfactory performance and mutual agreement.
Responsibilities
- Develop foundational security theory for agentic and multi-agent AI systems, as part of the research programme in multi-agent security.
- Build formal models of security-relevant behaviour in interactive systems and derive rigorous results that clarify key limits, trade-offs, and conditions under which stronger guarantees are possible.
- Run targeted computational experiments (e.g., small-scale simulations or proof-of-concept implementations) to validate assumptions and connect formal insights to realistic agentic settings.
- Develop and analyse formal models; derive provable guarantees, impossibility results, and trade-offs; design attack–defence formulations and evaluation protocols.
- Publish research in leading conferences and journals; contribute documented software to the group library; and support the supervision of graduate and undergraduate research projects.
- Assume substantial ownership over one or more core research thrusts within the lab’s multi-agent security programme.
Qualifications
- PhD/DPhil completed, or near completion, in theoretical computer science, information theory, cryptography, security, game theory, or a closely related area (ML acceptable with strong theory depth).
- Strong mathematical maturity with evidence of producing rigorous, security-relevant theory.
- Sufficient programming fluency (e.g., Python; familiarity with common ML tooling) to run computational experiments.
- Excellent communication skills, including writing for publication and presenting results.
Enquiries
Informal enquiries may be addressed to Dr Christian Schroeder de Witt (email: christian.schroeder@eng.ox.ac.uk)
Application details
Only online applications received before midday on 2 March 2026 can be considered. You will be required to upload a covering letter/supporting statement, including a brief statement of research interests (describing how past experience and future plans fit with the advertised position), CV and the details of two referees as part of your online application.
The Department holds an Athena Swan Bronze award, highlighting its commitment to promoting women in Science, Engineering and Technology.
Keywords
Multi-Agent Security, AI Security, Information Theory, Cryptography, Game Theory
Salary
£39,424 to £47,779 per annum: Grade 7
Postdoctoral Research Assistant in AI + Security (Early Career Fellowship) in Oxford employer: University of Oxford
Joining the Oxford Witt Lab for Trust in AI offers an exceptional opportunity to engage in cutting-edge research within a collaborative and innovative environment. The Department of Engineering Science fosters a culture of inclusivity and support, providing ample opportunities for professional growth and development, while being located in the vibrant academic hub of Central Oxford. With access to world-class resources and a commitment to advancing knowledge in AI security, this role is ideal for those seeking meaningful contributions to the field.
StudySmarter Expert Advice🤫
We think this is how you could land Postdoctoral Research Assistant in AI + Security (Early Career Fellowship) in Oxford
✨Tip Number 1
Network like a pro! Reach out to folks in the AI and security fields, especially those connected to the Oxford Witt Lab. A friendly chat can open doors and give you insights that a CV just can't.
✨Tip Number 2
Show off your skills! Prepare a portfolio of your past research or projects that highlight your expertise in multi-agent systems and security theory. This will help you stand out during interviews.
✨Tip Number 3
Practice makes perfect! Get ready for technical interviews by brushing up on your programming skills and theoretical knowledge. Run through potential questions with a friend or mentor to build confidence.
✨Tip Number 4
Apply through our website! Make sure to submit your application online before the deadline. Tailor your cover letter to reflect how your research interests align with the lab's goals—this shows you're genuinely interested!
We think you need these skills to ace Postdoctoral Research Assistant in AI + Security (Early Career Fellowship) in Oxford
Some tips for your application 🫡
Craft a Compelling Cover Letter:Your cover letter is your chance to shine! Make sure to highlight your research interests and how they align with our work at the Oxford Witt Lab. We want to see your passion for AI and security, so don’t hold back!
Tailor Your CV:When applying, tailor your CV to showcase relevant experience in theoretical computer science, cryptography, or any related fields. We’re looking for strong mathematical maturity, so make sure to emphasise your skills and past projects that demonstrate this.
Show Off Your Communication Skills:Excellent communication is key! Include examples of your writing for publication and any presentations you’ve done. We want to see how you can convey complex ideas clearly, so don’t forget to mention any teaching or mentoring experiences too.
Submit Before the Deadline:Don’t leave it to the last minute! Make sure to submit your application online before midday on 2 March 2026. We want to see your application, so get it in early and avoid any last-minute hiccups!
How to prepare for a job interview at University of Oxford
✨Know Your Research Inside Out
Make sure you can discuss your past research and how it relates to the role. Be prepared to explain your methodologies, findings, and how they connect to multi-agent security. This shows your depth of knowledge and passion for the field.
✨Brush Up on Technical Skills
Since programming fluency is key, ensure you're comfortable discussing Python and any relevant ML tools. You might be asked to solve a problem or explain a concept on the spot, so practice articulating your thought process clearly.
✨Prepare for Theoretical Questions
Expect questions that test your understanding of security theory, game theory, and cryptography. Review key concepts and be ready to discuss their implications in real-world scenarios, especially in relation to agentic AI systems.
✨Showcase Your Communication Skills
As publishing and presenting are part of the role, practice explaining complex ideas in simple terms. You might be asked to present your research or discuss how you would communicate findings to a non-technical audience, so be ready to demonstrate this skill.
