PhD Data Generation and User Simulation Research Intern — Fall 2026

Details

Today, NVIDIA is tapping into the unlimited potential of AI to define the next era of computing. The role sits on a research team focused on artificial data creation across pre-training, post-training, and evaluation infrastructure. Workstreams include population-grounded user simulation (synthetic users interacting with LLMs, calibrated against real behavioral signatures), verifier-grounded trajectory synthesis, multilingual and low-resource coverage, and SDG quality measurement across pre- and post-training corpora. The team measures success by downstream model performance (accuracy, robustness, calibration, multilingual parity, agentic safety) rather than by surface plausibility.

Responsibilities

• Research innovative techniques in generative models, artificial data creation, user simulation, reward modeling, and data-quality estimation for LLM training.
• Design and apply methods for high-fidelity synthetic data (e.g., behavioral calibration of simulated users, procedurally generated probes and scenario coverage, trajectory generation guided by verification, process-reward extraction from multi-step interactions, population-aware data mixing for pre- and post-training).
• Conduct experiments to validate that synthetic data measurably improves downstream model performance (accuracy, robustness, calibration, multilingual parity, agentic safety).
• Collaborate with researchers and engineers to integrate methods into production training and evaluation pipelines.
• Prepare research findings for internal presentations and potential publication at top-tier AI conferences.

Requirements

• Currently pursuing a PhD in Computer Science, Machine Learning, Computational Linguistics, Computational Neuroscience, or an equivalent program, with a specialization in deep learning, NLP, or LLM training.
• Research experience in at least one of: generative modeling, synthetic data generation, LLM post-training (SFT/RLHF/DPO/RL), reward modeling, multi-agent or interactive simulation, behavioral or cognitive modeling, or large-scale data curation.
• Excellent Python programming skills.
• Hands-on experience with deep learning frameworks (PyTorch) and the modern LLM training/serving stack (e.g., HuggingFace, vLLM, distributed training).
• Strong research background with publications at top-tier AI, ML, or NLP conferences.

Ways to stand out

• Experience training or fine-tuning LLMs end-to-end and evaluating them against real downstream tasks.
• Prior work on LLM-as-judge calibration, inter-rater agreement, or evaluator robustness for subjective dimensions.
• Prior work on user simulation, agent–user interaction modeling, or behavioral modeling grounded in real population data or cognitive science.
• Interest or background in multilingual / low-resource / sovereign-AI evaluation and training.
• Contributions to open-source projects in the SDG, LLM training, or evaluation space.

Compensation and benefits

• Internship hourly rate: 30 USD - 94 USD.
• Eligible for NVIDIA intern benefits (link provided in original posting).

Application and other details

• Applications accepted at least until May 26, 2026.
• This posting is for an existing vacancy.
• NVIDIA uses AI tools in its recruiting processes.
• NVIDIA is an equal opportunity employer committed to diversity and non-discrimination.