Most doctoral programs leave research skills to osmosis. You are expected to absorb version control, peer review, reproducibility and proposal writing from the air around you — and the gaps surface at the worst possible moment. A student who cannot reproduce their own Chapter 3. A submission that stalls on formatting. Here, each of those skills is explicit, demonstrated once on your own dissertation work, and then yours permanently.
Twenty-eight skills in six milestones. You choose what to work on and when — nothing is locked, and a first-year who needs the review cycle can take the review cycle. Each skill is performed as one week's ordinary research-log entry, so it replaces work rather than stacking on top of it. Two milestones a semester finishes the set in three to four semesters, which lands around when you propose.
The five surfaces you actually work on, plus the habit that makes all of them safe: checking a result against something you knew independently.
The test at the end is not a file listing. It is that another student ran your code on their own machine and got your number.
Evaluating somebody else’s argument, finding where the field stops, spotting an inference that fails, building a claim of your own, and saying it no more strongly than your evidence allows. This is where critical thinking is actually taught rather than assumed.
Everything that goes into a document somebody else can read, trust and cite — including alt text and colour-independent figures, which this institution needs in place by April 2027.
You referee a labmate’s draft and they referee yours; each of you then answers the review you received. Nothing here is simulated.
Four things that outlast you here: a citable deposit, a dataset somebody can still find the terms for in four years, an idea written well enough to be funded, and a person who learned to do research because you scoped their first task properly.
Every week you keep a research log in three channels, and it is graded every week. At the end of a semester you read back across the whole thing — which is the point, because progress in research is invisible at the scale of a single week and obvious at the scale of fifteen.
One paper, chapter, or set of conference papers. An annotated entry that goes past summary to what the evidence actually supports.
A page. Draft prose, an outline, a plan, a reflection — anything that moves the research forward on the page rather than in your head.
What you computed, what broke, what you are doing next. Where the code, the models and the solver work get written down.
The lab runs three tiers — doctoral research, master's team projects, and undergraduate REU and UROP students. By your third year you are supervising, which is why mentoring is a graded skill here and not a thing you are assumed to pick up.
What we do not require: prior optimization coursework, a background in energy systems, or having touched a terminal before. Those are what the first two milestones are for.
What the degree does require that you might not expect: one semester in an internship, taken as IE 5191 or an equivalent approved by the Graduate Advisor and your supervising professor. If you already have official project-based experience lasting about a semester, it can be waived, case by case.
Email Dr. Erick C. Jones Jr. with a paragraph about what you want to work on and why here. A CV is welcome; a cover letter that could have been sent to any lab is not useful to either of us.
Every lesson above exists in full — the reading, the mechanic, and the exact checklist it is graded against — and all of it is public at github.com/sear-labs/teaching-phd-skills. Read it before you write to us. It will tell you more about how this lab works than this page can. The lab's published models and data are public too, at github.com/sear-labs, if you would rather judge us by the code.
Two pages there are not graded and are the ones to read first. SEAR Lab Domain Knowledge is what we actually study — the three layers against the three research areas, with every cell filled in. It is the clearest statement anywhere of what you would be joining. Engineering Research Domain Knowledge is the wider discipline: what counts as evidence when you cannot run an experiment, and why a discount rate is a claim about how much future people matter.