Skip to content
Qurbit

Qurbit Research / Work

The projects came first, and the seats came after.

Qurbit is a research group that opens places to students. Which way round that goes decides most of what follows.

Finished
7 projects
Ongoing
45+ projects
Areas
3
Student seats
Per cycle
Frontispiece. Lorenz attractor, traced continuously

1Why the ordering matters

A program built to sell placements needs topics it can hand out on demand. Those topics have to be assignable to any student who pays, finishable inside a fixed term, and safe enough that the result can be called a success either way. Very little real research satisfies all three conditions at once, so the program invents work that does.

Because our projects exist whether or not a student joins them, they are allowed to be hard and they are allowed to fail. A student who spends a cycle establishing that a promising approach does not work has done something useful, and the writeup says so in those words. Negative results are also the easiest thing for a student to write honestly, since there is nothing to oversell.

The cost of this arrangement falls on applicants. Seats are capped by what the projects can absorb, not by how many good students want one, and in a given cycle that number is small. It is the part of this we would change first if we could.

Sold as a placementStudent paysTopic inventedWork assignedCertificateHow this runsProject existsSeat opensStudent joinsResult, either way
Figure 1. The same four steps in two different orders. Which one comes first determines whether the topic has to be safe.

2Areas

01Computer Science

Algorithms, systems, and computational methods, including work that leans heavily on implementation and measurement.

  • Algorithms and complexity
  • Machine learning
  • Software systems
  • Quantum computing
  • Data and simulation

02Finance

Markets, risk, and quantitative methods, with an emphasis on questions that can be tested against real data.

  • Quantitative modeling
  • Market microstructure
  • Risk and portfolio analysis
  • Financial data science
  • Behavioral finance

03Medical

Health, biology, and clinical questions, approached through data analysis and careful study design.

  • Biomedical data analysis
  • Public health studies
  • Computational biology
  • Medical imaging methods
  • Clinical research design

3Where the work ends up

Finishing a project and publishing one are separate events, and we keep them separate when we talk to students about it. Table 1 is roughly how a cycle turns out.

OutcomeWhat it meansFrequency
Internal writeupA complete, reviewed project that stays in house.Most cycles
Poster or abstractSubmitted to a student symposium or a conference student track.Some
Workshop paperA short paper, where the contribution carries one.A few
Negative resultThe approach did not work, documented well enough to save the next person the time.Any of the above
Table 1. Outcomes of a completed cycle. The first row is the common case, and it is a real one.
Completed cycleInternal writeupmost cyclesPoster or abstractsomeWorkshop papera fewa negative result can take any of the three
Figure 2. The same outcomes as a branching. Every route below is a finished project; only some of them leave the building.

None of those rows are hypothetical. Work from our projects has been accepted at NeurIPS, IEEE IROS, AAPM&R, APHA, ACSM Annual Meeting 2026, Annual Southeastern Pediatric Research Conference, International Conference on Emerging Infections and Public Health, and Preventive Medicine Reports, with more submissions under review.

We will not send weak work to a venue that accepts everything. Predatory conferences will publish a student for a fee within a week, and the line it buys on an application is worth less than nothing to anyone who looks it up.Admissions readers at selective schools do look these up. It is not a hypothetical risk.

4What a student actually does

A first placement is a defined piece of a larger question: a literature gap to map, a dataset to clean and characterize, a set of experiments to run, a published baseline to reproduce. The scope is set so that finishing is realistic in a cycle and so that the piece matters to the project, since work invented to keep a student busy is obvious to everyone including the student.

Map a literature gap01Clean and characterize a dataset02Run a set of experiments03Reproduce a published baseline04one of these, not all four
Figure 3. The four shapes a first placement usually takes. You get one of them, chosen to fit what you already know.

Students who come back for a second cycle usually take on the framing rather than an assigned question. A few have run their own thread with someone else helping.

5Read before you start

None of this is required to apply. All of it makes week one easier, and every accepted student gets pointed at the first two before their orientation call.

ReadingAuthorWhy we assign it
How to Read a PaperS. KeshavThree passes through a paper instead of one slow drown.
You and Your ResearchRichard HammingA talk on doing work that matters, by someone who did.
The Craft of ResearchBooth, Colomb & WilliamsFrom a question you like to an argument a stranger can follow.
Whitesides’ Group: Writing a PaperGeorge WhitesidesStart from the outline. The writing is not a report on the research; it is part of it.
Table 2. Assigned in roughly this order. The Keshav paper takes twenty minutes and pays for itself the same week.
Apply for a placementRead how a cycle runs