← Students

Advising Expectations for You and for Me

Last updated October 2026

What Makes a Great PhD Candidate?

Consistent Effort

In my experience, the students who succeed in a PhD are not necessarily the most brilliant or the most technically gifted, but the ones who put in consistent effort week after week. A very well-respected professor at a top engineering school once said to me: “If a PhD student can commit to 40 hours of research a week every week for 4–5 years, I can guarantee them a top 10 placement.” That’s a bold statement, but it contains a lot of truth.

The truth lies in two ideas: consistency (“every week for 4–5 years”) and a focus on research (“40 hours of research”).

First, consistency isn’t the same as intensity. Most PhD students work when they are inspired and cram all night before their weekly advisor meeting.

That’s not consistency. All-nighters result in exhaustion (where you’ll lose the next couple of days to recovery) rather than substantive insight.

Real consistency entails incremental progress and a pace you can sustain throughout the week for the long haul. Importantly, consistency often yields both higher productivity and more free time than irregular intensity. When I can get to the office early and (force myself to) leave by 5 or 6 pm for a few weeks in a row, I end up working MORE total hours per week and have MORE free time for a healthy life outside work. Try it yourself!

Second, what does it mean to “focus on research”? Not everything you do in graduate school is research. For example:

To be clear, you still need to do the things on the left. But they’re not research. The things on the right all count as genuine research, even though they might not seem like it.

In particular, a lot of academic research isn’t about “solving the problem” but rather about disseminating knowledge and making impact. Depending on your taste, you might find those parts of your PhD less “fun”, but expect to spend considerable time on them if you want to succeed.

Early in your PhD, when coursework is heavy and your dissertation topic is still forming, you may not reach 40 research hours each week. But increase your allocation gradually, and by the start of your third year, you should be at this level of consistent focus.

Curiosity

I believe the strongest predictor of research success is curiosity.

I believe the strongest predictor of research success is curiosity. Great PhD students

• attend seminars broadly (even when it’s not directly related to their current research),
• skim papers (both recent and foundational) every week,
• talk to peers about potential ideas and the challenges each of them is facing, and,
• most of all, recognize that true scholarship is voracious.

Consume books, podcasts, art, music, and learn everything you can, down to the history of school segregation in the US. Exposure to different ideas matters more than it might seem.

In the same vein, great PhD students go beyond the tasks outlined in a meeting. They think about why they are doing those tasks. (That’s the same curiosity at work.) If something doesn’t work, they figure out why it failed, because precisely articulating why something failed often suggests a solution. When something does work, they reflect on the implications and what to try next.

This might all seem obvious. But I can’t tell you how often I’ve seen a PhD student generate a plot for a meeting from a simulation, but not spend the time to look at it, interpret it, or think about what it’s telling them.

Choosing Projects

We’ll always aim for projects at the intersection of our interests. I’ll play a large hand in helping choose your first project, to maximize its chance of success. As you progress through the PhD, I expect you to lead the research.

How We’ll Run Our Meetings

We will meet one-on-one (or two if you have a co-advisor) every week. Come prepared with:

• a short write-up of progress, or
• for a detailed proof, a draft sent a few days in advance so I have time to read it.

Meetings are not just for reporting — they are for clarifying next steps. With about 10–15 minutes left, if we haven’t identified concrete tasks for next week, stop and ask: “Can we clarify what the specific to-dos are for next time?” I expect you to play an active role here: suggest next steps, and I will guide or redirect if needed.

Weekly minutes. Take notes during our meeting, then email me short minutes and action items afterward. I’ll often start our next meeting from that email.

There are no weeks when “nothing happened.”

Finally, there are no weeks when “nothing happened.” Think carefully about failed experiments and dead ends in proofs, and describe them and what they suggest. Even if you spent the week preparing for midterms, you should be able to tell me about what you learned, what you found cool, what was difficult, etc.

When you’re confused, email me early. Don’t delay because you feel embarrassed about a lack of progress or a “silly” question — I would rather hear from you early than lose a week to silence.

I often work from the office, so feel free to come knock on my door with a cool idea. There’s no requirement for you to work from the office, but I do expect you to be in LA and generally available for on-campus meetings with notice.

What You Can Expect From Me

• Protected time. My students are a top priority, and I will do my best to protect our weekly meeting time for you.
• Responsiveness. I’ll try to respond to emails within one business day.
• Feedback. For written feedback, I usually need a day or two to respond to a short proof and up to a week for a large section of a paper. Around deadlines we jointly commit to, I will prioritize same-day feedback, increased responsiveness as the deadline approaches, and make myself available. I expect the same effort from you, so we can submit strong work on time to high-quality venues.
• Conferences. As you advance beyond the first year of the program, we will plan which conferences you attend to present the results of your research, and I’ll help you find and prepare for these opportunities.
• Vacation. I’ll give you a heads-up about my own time away, too.
• Support. If you feel comfortable, talk to me about the personal challenges of a PhD. I have been there, and part of my role is to help you find your way through it.

Using AI

AI tools are changing how research gets done, and how I work with them is still evolving. These norms will evolve too. For now:

• Use AI tools freely. I do, every day.
• You own everything you bring to a meeting. If you can’t explain it without the tool, it isn’t done yet.
• Expect me to ask you to walk me through a derivation, a piece of code, or a result in your own words, often at the whiteboard.
• Tell me when AI substantially produced something: a proof idea, a block of code, a draft paragraph.

Authorship and Collaboration

In our field, authorship is alphabetical unless otherwise agreed. If you would like to add an additional co-author — especially a faculty member — discuss it with me before reaching out. This ensures clarity about contributions and avoids awkward misunderstandings.

Internships and Careers

Summer internships are strongly encouraged. Seeing academia and industry side by side (different work environments, research problems, and avenues for impact) is hugely important as you begin your career.

While many of my past students have gone on to academia, I also welcome students targeting industry. Some of the biggest impact you can have is outside academia, and I spent five years as a quant at Barclays before choosing academia. Whatever comes next, what I look for is a genuine commitment to research while you’re here. That’s what makes the work good, and it’s what makes you valuable wherever you go.

Taking Care of Yourself

Take at least four weeks of vacation each year. Academia slows down in summer and winter, so those are often good times. Just give me a heads-up so I can plan.

A PhD is emotionally demanding. Burnout, impostor syndrome, and perfectionism are real. Wins are rare, and the work is always uphill. Nicheness of topics can make research feel isolating. Prioritize your mental health at all times.

Develop hobbies, friends, and pastimes in LA outside of work.

Building Your Community

Senior PhD students are great resources for things like filling out forms, LaTeX, or computing resources at USC. Build relationships with your peers: talk about research, life, or just whinge about me as your advisor. A supportive peer network is one of the best ways to thrive in graduate school, and establishing your professional network now is one of the highest dividend-paying investments you can make for the rest of your career.

Finally, my group values kindness and respect. Be generous with feedback and ideas.

On that topic: I’m a new dad with a little one at home. Unexpected things will come up. I hope to navigate it all with compassion and understanding, and expect the same in return.

Acknowledgements

This document draws heavily on Advising Expectations and Commitments (2025), which I co-wrote with Andrew Daw. That document was inspired and informed by similar documents from Angela Zhou, Sean Sinclair, and Chara Podimata, with thanks to André Calmon and Adam Elmachtoub for feedback on early drafts.