Joining the Lab

Postdoctoral Fellows  •  PhD Students  •  Visiting Researchers  •  Undergrad & Master's Researchers

Postdoctoral Fellows

Fields Institute-Principles of Intelligence Postdoctoral Fellowship: Formal Methods for Frontier AI Systems

Applications are invited for a two-year Fields Institute-Principles of Intelligence Postdoctoral Fellowship to conduct research on formal methods for the safety and security of frontier AI agents and multi-agent systems. Research directions of interest include, but are not limited to, temporal logics over finite traces for specifying frontier AI agent behaviour, monitor synthesis and runtime enforcement for actions that are not permitted, and quantitative and probabilistic guarantees for stochastic policies for frontier AI systems. Candidates are encouraged to interpret these research directions broadly. Successful applicants will be strong technically and have an interest in safety and security of frontier AI systems. We are looking for applicants with strong research skills, ideally with multiple publications at top venues in verification, AI safety, and machine learning (e.g., CAV, SAIV, Automatica, NeurIPS, ICLR, ICML, AISTATS). Candidates must have a PhD or equivalent degree in computer science, mathematics, statistics, or a closely related field. Strong programming skills are required; experience with LLMs is welcome but not required.

Fellows will have access to mentorship from experts in verification, security, and AI safety at the Fields Institute, the University of Toronto, the Vector Institute for Artificial Intelligence, the Schwartz Reisman Institute for Technology and Society, and PrincInt. Women and underrepresented minorities are especially encouraged to apply. Applications are reviewed on a rolling basis and the posting will remain open until all positions are filled. The salary range is CAD $70,000–$90,000 per year. For informal enquiries before applying, contact Yevgeny Liokumovich ([email protected]) or Tim G. J. Rudner ([email protected]).

We are looking for postdoctoral fellows to start as soon as possible.

The Fields Institute is a centre for mathematical research activity - a place where mathematicians from Canada and abroad, from academia, business, industry and financial institutions, can come together to carry out research and formulate problems of mutual interest. Our mission is to provide a supportive and stimulating environment for mathematics innovation and education. Learn more about the Fields Institute here.

Application Process

Applications can be submitted via this form. Applicants must submit all of the following documents:

  1. Curriculum vitae
  2. Two representative publications (preprints are acceptable)
  3. Statement of research (2 pages) describing prior research experience and future research plans
  4. Link to Github account

2-3 letters of recommendation will be solicited after the initial review. Applicants are encouraged to email me after submitting their application.

≫ Status: Applications Open (rolling review)

PhD Students

There are two main avenues through which I advise PhD students: PhD programs at the University of Toronto and DPhil programs at the University of Oxford.

This year, I plan to recruit:

  • 2 students through the PhD program in Statistics,
  • 1 student through the PhD program in Computer Science, and
  • 1 student through Oxford.

Note: Prospective applicants are highly encouraged to apply for external funding. Receiving external funding significantly increases the chances of admission. For a list of external funding sources, see this spreadsheet.

≫ PhD expression of interest: Fill out this form if you want to let me know you’re thinking of applying.

Option 1: PhD at the University of Toronto

I’m an Assistant Professor at the University of Toronto, and I hold appointments in both the Department of Statistical Sciences and the Department of Computer Science. I can advise students in either department.

I’m also a faculty member of the Vector Institute for Artificial Intelligence, and all of my students are automatically affiliated with the Vector Institute.

Note: I highly recommend applying to the PhD program in Statistics if you have a strong mathematics background and if any or all of the following applies to you: (i) you have Canadian citizenship or (ii) you are potentially interested in theory research.

PhD in Statistics

For information about the PhD in Statistics, see here.

If you want to select me as a potential advisor, please specify my name on the application form. I am generally open to co-advising students with other faculty members in the Department of Statistical Sciences or other departments.

Timeline: Applications will be reviewed in December to January. Interviews with shortlisted candidates will be conducted in January to February.

≫ Deadline: November 22, 2026

≫ After you apply: Fill out this form to let me know you applied.

PhD in Computer Science

For information about the PhD in Computer Science, see here.

If you’re applying to the PhD program in Computer Science at the University of Toronto and would like to select me as a potential advisor, please select “Other” on the supervisor list and specify my name.

Timeline: Applications will be reviewed in December to January. Interviews with shortlisted candidates will be conducted in January to February.

Note: Prospective applicants are encouraged to identify potential co-advisors in the Department of Computer Science. Faculty members with shared research interests include: Sheila McIlraith, Colin Raffel, Nicolas Papernot, Roger Grosse, Zhijing Jin, and David Duvenaud, among others.

≫ Deadline: December 1, 2026

≫ After you apply: Fill out this form to let me know you applied.

Option 2: DPhil at the University of Oxford

I’m an Associate Member of the Department of Computer Science and a Faculty Affiliate at the Oxford Martin School AI Governance Initiative (AIGI).

I co-advise DPhil students in the AIMS CDT as well as in the direct-entry DPhil programs in Computer Science and in Engineering Science. For applicants to the AIMS CDT or direct-entry DPhil programs, I highly recommend also applying for the Oxford AI Governance Initiative (AIGI) studentship and mentioning me as a potential co-supervisor.

Note: Prospective applicants are encouraged to apply through the AIMS CDT and mention me as a potential advisor in their application. Since I am primarily based in Toronto, I only co-advise students at Oxford, and prospective applicants must have a co-advisor alongside me. Students accepted into the AIMS CDT program can use the first year of the program to identify a co-advisor, but I recommend mentioning 1-2 additional Oxford faculty in the application. Oxford faculty members with shared research interests include: Phil Torr, Michael Osborne, Robert Trager, among others.

ELLIS PhD Program (also see Option 3 below): Oxford applicants who want to work with me are encouraged to also apply to the ELLIS PhD Program even if they want to join my group at the University of Oxford.

≫ Deadline (DPhil AIMS CDT): mid January, 2027

≫ After you apply: Fill out this form to let me know you applied.

Option 3: ELLIS PhD Program

I’m a member of the European Laboratory for Learning and Intelligent Systems (ELLIS) and advise students through the ELLIS PhD Program. Each ELLIS PhD student has two co-advisors at different institutions. The program includes a research exchange of at least six months at the co-advisor’s institution, as well as access to ELLIS summer schools, workshops, and the wider ELLIS network.

Students in this track are co-advised by me at the University of Toronto and by an ELLIS faculty member at a European institution. They spend part of their PhD in Toronto and part in Europe. This option suits applicants who want close ties to both the North American and European AI research communities.

To be considered, apply through the centralized ELLIS application portal and list me as a potential advisor. You must also apply to the PhD program in Statistics or Computer Science at the University of Toronto if you want me as your ELLIS home-institution advisor (see Option 1).

Note: Applicants should select European ELLIS co-advisors whose research interests are well aligned with mine. ELLIS members with shared research interests include: Phil Torr, Michael Osborne, Maksym Andriushchenko, Francis Bach, among others.

≫ Deadline (ELLIS PhD Program): October 31, 2026

≫ After you apply: Fill out this form to let me know you applied.

Visiting Researchers

I occasionally host visiting researchers from other institutions. I only host graduate students from other institutions and experienced researchers (with rare exceptions). The best way to work with me as a visitor is through the Vector Institute Research Internship Program.

Interested students are encouraged to email me to express their interest. The email should include a CV, a description of research interests, and links to 2-3 representative papers.

≫ Status: Applications Closed

Undergraduate and Master’s Research Assistants

I welcome expressions of interest from exceptionally well-prepared undergraduate and master’s students at the University of Toronto to contribute to research projects in my group. To make a meaningful contribution, a minimal commitment of 15h/week is expected and a commitment of 15-20h/week (or more) is preferred. Students do not need to have prior research experience but are expected to have strong programming skills and solid foundations in machine learning and mathematics and, ideally, in statistics.

≫ If you would like to work with my group: Please fill out this form.