Senior Ml Scientist • Senior

Who you are

You have a Ph.D. or equivalent research experience in Computer Science, Electrical Engineering, Mathematics, or a related field, and you bring over 5 years of experience delivering computer vision and generative AI models in production. Your portfolio includes first-author publications or patents in top-tier venues such as CVPR, ICCV, ECCV, SIGGRAPH, NeurIPS, or ICLR, showcasing your deep expertise in the field. You possess a strong command of PyTorch and CUDA, which you have utilized to build and optimize complex models. Your hands-on experience with advanced diffusion, transformer, and multimodal text-to-image/video models demonstrates your ability to innovate in generative AI.

You are skilled in developing and optimizing training and inference pipelines, employing techniques such as data parallelism, quantization, and distillation. Your familiarity with TensorRT and DataBricks allows you to enhance model performance and streamline workflows. You understand the importance of human-in-the-loop evaluations and are adept at creating side-by-side visual comparisons and calculating technical metrics. Your collaborative spirit shines through as you work closely with product, design, and engineering teams to ship research-driven features at scale.

Desirable

Experience tuning and training custom LLM models for AI assistants like Qwen, LLama, and GLM would be a significant advantage. You are also encouraged to apply even if your experience doesn't match every requirement, as we value curiosity and a growth mindset.

What you'll do

As a Senior ML Scientist at PicsArt, you will be at the forefront of inventing and deploying breakthrough generative AI capabilities that impact millions of creators globally. Your primary responsibility will be to build state-of-the-art tools that enhance visual expression. You will invent and publish advanced models, focusing on high-resolution generation up to 16K, and develop innovative features for image and video retouching, effects, quality enhancement, and avatar generation.

You will design, build, and maintain training and evaluation workflows, ensuring that they are efficient and effective. Your role will involve collaborating with cross-functional teams to integrate research-driven features into the product, ensuring that they meet the needs of users while maintaining high standards of quality. You will also be responsible for optimizing training and inference pipelines, leveraging your expertise in data parallelism and quantization to improve performance.

In addition, you will create human-in-the-loop evaluations, conducting side-by-side visual comparisons and calculating technical metrics to assess the effectiveness of the models you develop. Your work will not only shape the future of visual expression but also contribute to the overall mission of PicsArt to bring creativity to the world.

What we offer

At PicsArt, you will be part of a dynamic team that values creativity and innovation. We offer a collaborative work environment where your contributions will have a direct impact on the products we create. You will have the opportunity to work with cutting-edge technologies and be at the forefront of generative AI research. Our company is headquartered in Miami, with offices around the world, providing a global perspective on creativity and technology. We encourage you to apply even if your experience doesn't match every requirement, as we believe in the potential of diverse teams to drive innovation.