Research Fellow (Physics)

Interested applicants are invited to apply directly at the NUS Career Portal. Please note your application will only be processed if you apply via NUS Career Portal. NUS Career Portal link: https://careers.nus.edu.sg/job/Research-Fellow-%28Physics%29/33771-en_GB/?st=B73C88EDE62E3181170B6DCC690C7A5FD31BC3BD We regret that only shortlisted candidates will be notified. Job Description The successful candidate will work with Associate Professor Duane Loh on Electron Microscopy under a project on AI-assisted electron microscopy. About the role We are seeking a research staff member to anchor the experimental side of a closed-loop imaging programme that couples electron microscopy with in-house foundation models. The successful candidate will prepare and image low-dimensional materials, and will collect and log these datasets in a form that our symmetry- and context-aware models can learn from. The long-term goal is semi-autonomous, and eventually autonomous, microscope operation in which model inference guides acquisition in real time. Key responsibilities Prepare samples for transmission electron microscopy, including specimens supplied by collaborators. Operate (S)TEM instrumentation autonomously, including spectroscopic modalities across the beam such as EELS and EDS. Design and execute acquisition campaigns that yield datasets of sufficient volume and quality for model training, with particular attention to structural motifs that are functionally consequential. Maintain rigorous, structured logging of acquisition metadata so that datasets are model-ready rather than merely archived. Work with the computational members of the group to close the loop: feeding curated datasets into foundation-model development, and feeding model outputs back into acquisition strategy. Contribute to publications and collaborator reporting. Qualifications PhD in Physics, Materials Science, Chemistry, Chemical Engineering, or a closely related discipline; or an MSc with at least three years of hands-on microscopy experience in a research setting. Skills: Independent (S)TEM operation: column alignment, aberration correction workflows, astigmatism correction, dose management. Analytical modalities: EELS and/or EDS acquisition, including spectrum imaging and basic quantification. Sample preparation appropriate to low-dimensional materials — transfer methods, grid selection, cleaning protocols, and the judgement to know when a specimen is compromised rather than merely difficult. Practical understanding of beam damage, contamination, and drift, and how each constrains what can be asked of a sample. Capacity to work autonomously on a research programme with loosely specified intermediate steps. Clear communication across the experiment–computation boundary; the role sits at that interface and will fail if the person cannot operate in both registers. Reliability in handling collaborator samples, including realistic scheduling and honest reporting when a specimen does not yield usable data. Experience: Documented experience operating transmission or scanning transmission electron microscopes independently — i.e. without supervision for routine alignment, sample loading, and data collection. Doctoral-level research experience in materials science, physics, or a closely related field, with electron microscopy as the primary experimental method rather than an occasional supporting technique. Two or more years of independent (S)TEM operation, including alignment, troubleshooting, and unsupervised data collection sessions. Demonstrated experience with analytical EM, in particular EELS. Experience with momentum-resolved or spectrum-imaging modes is a strong advantage, as is any acquisition mode that produces high-dimensional data rather than single images. Sample preparation for low-dimensional materials, including focused-ion-beam (FIB) work. Experience with helium-ion microscopy (HIM), transfer methods, or other structure-definition techniques is valuable. Direct experience with 2D materials and/or nanostructured catalysts — enough familiarity with the systems to judge specimen quality and anticipate beam-induced artefacts rather than discovering them after the fact. Python and data-pipeline experience applied to their own experimental data: signal processing, batch analysis, and the construction of workflows that others can rerun. Track record of peer-reviewed publication, including work as a contributing author on multi-group collaborations. We t…