Graduate seminar · Department of Chemical Engineering, Chulalongkorn University
17 August 2026
A conference is not a reward
for work already finished.
It is an instrument you deploy
to answer a question.
Everything in the next fifty minutes follows from this one sentence.
Today
Interrupt earlier if something is unclear. The last ten minutes are yours either way.
Why this is an instrument
Whether someone across the world solved your problem eight months ago and has not published it yet.
Whether your interpretation holds when a stranger who has no reason to be kind attacks it.
Which people have the instrument, the sample, or the model you are missing — and would say yes.
If a meeting cannot answer at least one of these for the work you have right now, it is tourism.
Part 1 · 8 minutes
Match the venue to the stage the work is at, not to the prestige you would like.
Part 1
Early
Domestic or regional meeting, small session. The goal is criticism, not visibility — you want the awkward question now, while the design can still change.
Take the poster. Thirty minutes per person beats three.
Middle
Specialist international meeting in your subfield. The goal is to test the interpretation against people who know the system as well as you do.
Fight for the oral slot. This is where collaborations start.
Late
Large flagship meeting, or an invited talk. The goal is priority, visibility, and reviewer familiarity.
Least useful for improving the work. It is already fixed.
One question decides it: what would have to happen at this meeting for the trip to have been worth it?
Part 1
Warning signs — an invitation email that flatters you · a scope covering five unrelated fields · a fee due before review · a venue that changes city every year · no named programme committee
Part 2 · 8 minutes
Everything that decides the value of the trip is decided before you board.
Part 2
T−6 months Decide the claim Name the one sentence you intend to defend. Identify the experiment still missing to support it.
T−4 months Close the data gap Run the missing measurement. Assume one repeat will fail, and budget for it.
T−3 months Submit the abstract Promise the result you will have. Not the topic you are working on.
T−6 weeks Build the materials Figures at final quality. Draft slides or poster. Travel booked, paperwork started.
T−2 weeks Rehearse and target Two full rehearsals with questions. Write down the people and sessions you are going for.
The abstract is a contract, not a title. “The effect of X on Y was investigated” tells the committee you do not yet know what you found.
Part 2
1
Who you are, which group, what problem you work on. Rehearsed until it bores you. You will use it forty times.
2
Question, approach, key result, why it matters. No slides. This is what you say standing next to a coffee urn.
3
Title, one figure, three results, your email, a QR code to the preprint. People lose business cards; they keep paper with a figure on it.
4
Five people you want to meet. Three sessions you will not miss. One question you want answered. Written down, on your phone.
Part 3 · 10 minutes
A twelve-minute talk is not a short seminar. A poster is not a paper on a wall.
Part 3
One message per slide, written as the slide title · about one slide per minute · memorize only the first 60 seconds and the last 30 · finish 30 seconds early
Part 3
3 m
Three seconds, while walking past. Write the finding, not the topic. “Additive X triples zinc anode life” stops people. “A study of additives for zinc anodes” does not.
1 m
A visitor who stops gives you thirty seconds of silent reading. One large central figure, labelled so it works without you standing beside it.
arm’s length
Method, conditions and controls in small type at the edges. Present for the specialist, invisible to everyone else.
Stand beside your poster, not in front of it · open with “May I give you the two-minute version?” · put your name and email large enough to photograph
Part 4 · 9 minutes
Nobody in that room is grading your grammar. They are watching how you think.
Part 4
1
“What was the scan rate?” Easy. Answer in one sentence and stop talking.
2
“How do you know it is not an artefact?” Name the control you ran — or concede that you have not run it.
3
“Could this be X instead?” Often the most useful question you will get. Say what evidence would separate the two.
4
“Does this work for other systems?” Answer where the boundary is, not a hopeful yes.
5
Rarely hostile. Thank them, write the reference down where they can see you do it, and mean it.
Part 4
1 Acknowledge “Thank you, that is an important point.” Buys three seconds and costs nothing.
2 Restate “So you are asking whether the capacity loss comes from the anode rather than the electrolyte?” Lets them correct you cheaply.
3 Answer in one sentence The direct answer first. Not the background, not the history.
4 Offer the detail “I can show the control experiment if that is useful.” Let them choose to go deeper.
“Could you repeat the question, please?” · “I want to make sure I understood — do you mean …?” · “I do not know. What would settle it is …” · “That is a fair criticism, and here is what we did about it.” · “Could we continue this at the coffee break?”
Part 5 · 6 minutes
This is where most conference value is lost — quietly, and permanently.
Part 5
Write the trip note. Two pages on the plane, while it is fresh: what you learned, who you met, what changed in your plan.
Send the follow-ups. Three sentences each: where we met, what we discussed, what I propose next. Attach the one-pager. After a week they no longer remember your face.
Turn questions into work. Every question you could not answer becomes an experiment, a control, a paper to read, or a paragraph to rewrite.
Report back to the group. Fifteen minutes at group meeting is how one funded trip becomes useful to twelve people.
Send the follow-up emails from the airport. Waiting until you feel rested means waiting until it is too late.
If you remember three things
A conference is a research instrument. Before the next one you attend, write down what you intend it to measure.