Multimodal Vision Research Laboratory MVRL

Join the Lab

Recruiting postdocs, research staff, students, and visitors

We're looking for people with strong technical foundations who want to work on hard problems in computer vision and multimodal learning, spanning geospatial AI, biodiversity and conservation, generative modeling, and representation learning. See our recent publications for a sense of the kind of work we do.

Submit an Interest Form Takes ~5 minutes. We review on a rolling basis and typically reply only when there is a potential fit.

Please review our FAQ below before reaching out. Note: we are not currently accepting high school students—see this FAQ item for guidance. If you are applying to a listed opening below, use that position's application form rather than the general interest form.

Open Positions

Postdoctoral Researcher — Hybrid Neural Fields for Earth Observation

Now recruiting

Taylor Geospatial funded · In person, St. Louis · ~2 years, starting Summer 2026 or later (extensions possible)

Join a multi-institution project on next-generation neural-field representations of Earth observation data—continuous, queryable representations of the planet that can be updated as new observations arrive and that report calibrated uncertainty over user-specified regions and time ranges. Funded by Taylor Geospatial, with engagement on the broader Features of the World initiative. Strong publication record at top-tier computer-vision and machine-learning venues (CVPR, NeurIPS, ICLR, ICCV, ECCV) required.

Full Position Details Apply for This Opening

Why Join MVRL

We publish regularly at top-tier venues (CVPR, ICCV, ECCV, NeurIPS) and work on interdisciplinary problems with real-world impact. Day-to-day work centers on multimodal representation learning, geospatial AI, biodiversity and conservation, and generative modeling over earth observation data—usually in close collaboration with other students, postdocs, and external partners.

  • Top-venue computer vision & ML
  • Planet-scale geospatial data
  • In-person lab culture in St. Louis
  • Mentorship from PhD students & postdocs
  • Cross-institution collaborations
  • Path to strong PhD programs & industry roles

Frequently Asked Questions

Expand the question that matches your situation.

The specifics depend on the role, but in general we look for:

  • Strong programming skills, primarily in Python.
  • Solid foundations in machine learning and computer vision, with hands-on experience using PyTorch or a similar framework.
  • Mathematical maturity in linear algebra, probability, and optimization.
  • Clear written and oral communication—research only matters if it can be shared.
  • A collaborative spirit. Most of our work involves close collaboration across students, postdocs, and external partners.
  • For PhD and postdoc applicants, a track record (or strong potential) for publishing at top venues such as CVPR, ICCV, ECCV, or NeurIPS.

To save everyone time, these inquiries are usually a poor match:

  • Remote-only arrangements (lab work is in person in St. Louis).
  • Little or no machine learning / computer vision experience.
  • Agendas that only overlap our older or inactive application areas, without a clear method-level connection to current work (see research areas and recent publications).
  • Generic mass emails that do not engage our papers, openings, or people.

The interest form above does not replace a formal PhD application. Prospective students apply through the WashU PhD program in Computer Science, Imaging Science, or Electrical & Systems Engineering; deadlines are typically in mid-December. We recommend attending the appropriate recruiting Info Session in the fall, and you are welcome to also submit the interest form so we know you are applying.

Listing Nathan Jacobs as a faculty of interest is enough for us to see your application—you do not need to email directly. Cold emails are rarely answered unless there is a unique connection (uniquely well-aligned research interests, prior work with someone we know, or a similar concrete link).

If a matching opening is listed above, use that position's application form. Otherwise, submit the general interest form. The form has fields for your funding situation (including external fellowships, home-institution support, or self-funded visits), so use those to tell us what kind of arrangement you have in mind. These roles are in person in St. Louis.

You're welcome to submit the interest form above, and we encourage it. In parallel, we typically defer to our PhD students to choose who they collaborate with on day-to-day projects—so it helps to do some homework first. Look through our research areas, identify one or more PhD students whose work overlaps with your interests, and name them in your interest form (or reach out to them directly). Each PhD student keeps recent publications on their personal website, which is the best way to see what they're currently working on.

Applications are much more likely to find a match when they include a concrete idea for a project you'd want to work on with a specific student. It doesn't need to be long, but showing how your interests align with their ongoing work goes a long way.

We generally expect undergraduate researchers to commit around 10 hours per week during the semester. Most of that time is spent in or near McKelvey Hall, where the lab is based. During the academic year, most students participate as volunteers or through an independent study for course credit (e.g., CSE 4001). We occasionally offer paid positions in the spring and fall, but openings are limited and infrequent. Sustained semester-long involvement with regular access to lab space typically requires one of these formal arrangements; once you've matched with a mentor, they can help you figure out what fits.

Yes—several WashU-administered programs place undergraduates with research labs (including ours) for the summer. Applications typically open in December and close in mid-February; check each program page for exact dates.

For these programs, you must apply directly through the program itself—just mention your interest in working with MVRL in your application.

We do not currently take on high school students. The best preparation is to build a strong foundation in machine learning and computer vision—work through introductory courses, get comfortable with PyTorch, and read recent lab papers on topics that interest you. Once you're an undergraduate with some experience, please feel free to reach back out.