Key Responsibilities

• Research, evaluate, and benchmark generative and diffusion-based models (Stable Diffusion, Sora-like models, GANs, NeRFs) for simulation and synthetic data generation.
• Build pipelines to replicate images and videos across new environments, lighting conditions, scenes, poses, and object variations.
• Develop multimodal prompt-based simulation workflows including:
  • Text → Image
  • Image → Image
  • Video → Video transformations
• Fine-tune models for domain-specific simulation tasks such as:
  • Texture transfer
  • Background replacement
  • Camera simulation
  • Noise injection
  • Motion variation
• Create automated pipelines to scale image, video, audio, and text simulation across large datasets.
• Evaluate realism, fidelity, annotation consistency, and domain-adaptation effectiveness of generated synthetic data.
• Work closely with ML researchers to integrate synthetic data into training loops and improve downstream model performance.
• Collaborate with backend and data teams to design scalable storage, sampling, and dataset versioning strategies for simulation workflows.
• Develop metrics and QA processes for simulation quality, drift detection, and dataset reliability.
• Support training pipelines, experiment tracking, and dataset versioning as simulation infrastructure scales.

Preferred Experience

• Experience with multimodal generative models for image, video, and text-prompted generation.
• Familiarity with dataset versioning and experiment tracking tools such as DVC, Weights & Biases (W&B), or MLflow.
• Understanding of domain adaptation and synthetic-to-real generalization techniques.

Qualifications

• 3–6 years of experience in applied machine learning or generative AI.
• Strong Python programming skills with hands-on experience in PyTorch or TensorFlow.
• Practical experience working with generative models including diffusion models, GANs, video synthesis models, and NeRFs.
• Familiarity with data augmentation, image/video transformations, and synthetic data generation workflows.
• Experience building scalable pipelines using FastAPI, Airflow, or custom orchestration frameworks.
• Understanding of GPU-based training, inference optimization, and model performance tuning.
• Practical knowledge of Git, Docker, Linux, and cloud platforms such as AWS, GCP, or Azure.