Applied Scientist Intern: Differentially Private Video Synthesis

Job Description

Overview

Join a strong research team advancing high-fidelity synthetic data generation pipelines. You will work on cutting-edge research in differentially private synthetic data generation, with a particular focus on video generation. Opportunities include advancing our existing work on Private Evolution and Differentially Private Reinforcement Fine-tuning, designing new innovative algorithms for DP synthetic data, and applying them to high-impact real-world projects.

Responsibilities

• Advance the research: Develop cutting-edge algorithms for differentially private video synthesis, including mechanisms, training objectives, and/or optimization strategies that improve fidelity, stability, and downstream utility.

• Own experiments end-to-end: Define hypotheses, build rigorous evaluation protocols for synthetic video data (e.g., classification, detection, pose/landmarks, segmentation, tracking), run ablations, and analyze results at scale.

• Collaborate broadly: Work closely with researchers and engineers to co-design algorithms, benchmarks, and experimental setups aligned with broader program goals.

• Communicate impact: Present findings clearly, contribute to internal reports, and—where appropriate—author submissions to top-tier research venues.

Qualifications

Rquired/Minimum Qualifications:

• Currently enrolled in a PhD program (senior stage), or a Master’s student with a strong publication record.

• Peer reviewed publications or preprints in top venues (e.g., ICCV, CVPR, SIGGRAPH, ECCV, NeurIPS, ICLR) or equivalent evidence of research impact.

• Research experience in image and video synthesis and deep learning, including training, fine tuning, and evaluation of modern vision models.

• Strong skills in experimental design, quantitative analysis, and clear scientific communication.

Preferred/Additional Qualifications:

• Experience with models or agents for video generation.

• Knowledge of human-centric tasks (pose/landmarks, tracking, activity recognition, face/hand analysis).

• Demonstrated ability to collaborate in multidisciplinary teams and to drive projects with minimal supervision.

This position will be open for a minimum of 5 days, with applications accepted on an ongoing basis until the position is filled.

Microsoft is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to age, ancestry, citizenship, color, family or medical care leave, gender identity or expression, genetic information, immigration status, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran or military status, race, ethnicity, religion, sex (including pregnancy), sexual orientation, or any other characteristic protected by applicable local laws, regulations and ordinances. If you need assistance with religious accommodations and/or a reasonable accommodation due to a disability during the application process, read more about requesting accommodations.