Jobs › Argonne National Laboratory
Postdoctoral Appointee - AI/ML for Particle Accelerators
Posted Jul 22, 2026 · We last checked this listing on Oct 6, 2026
Apply at Argonne National Laboratory
Likely interview questions for this role
Written from this job description, not a generic list. Each one notes what the interviewer is really checking.
About your past
Tell me about software you wrote that others relied on. How did you make it dependable?
What they are checking: Whether they have the strong programming skills the posting asks for, including testing and maintenance.
Tell me about a time you had to convince skeptical experts to try a new approach.
What they are checking: Whether they can earn trust from experienced operators and physicists who are cautious about change.
What would you do
Your PhD may not be in accelerator physics. How would you get up to speed quickly?
What they are checking: Whether they have a realistic plan to learn the domain that the posting says can be new to them.
You get a short window of machine study time and your method doesn't work as expected. How do you use the rest of the shift?
What they are checking: Whether they can adapt experiments in real time and still learn something from limited beam time.
About the work
How would you use an LLM-based agent to help an operator diagnose a fault in the APS injector?
What they are checking: Whether they can picture a concrete, useful role for agentic AI in a real control room.
Where would you not want an AI agent making changes to a particle accelerator, and how would you put guardrails in place?
What they are checking: Whether they think seriously about safety and trust when AI touches machine control.
What machine learning methods have you used for tuning or optimization, like Bayesian optimization or reinforcement learning?
What they are checking: Whether they know the methods most used for accelerator tuning and their tradeoffs.
Explain how beam emittance or orbit is measured and why it matters for a storage ring.
What they are checking: Whether they have the accelerator physics basics or can show they are picking them up.
How would you measure whether an AI tool actually helps operators make better decisions?
What they are checking: Whether they think about evaluation from the user's side, not just model accuracy.
How would you handle noisy, drifting data from accelerator diagnostics in your models?
What they are checking: Whether they understand real machine data problems like sensor drift and changing conditions.
Practice this interview out loud.
Offer builds a real interview for this exact role at Argonne National Laboratory from your resume and this job description, asks the questions one at a time, and tells you what landed. The first one is free.
Practice this out loudThe full job description
As published by Argonne National Laboratory.