Creating a poster from a lengthy document can be a tedious task. Imagine having an AI assistant that could automatically summarize the key information and design a visually appealing poster for you. Researchers have developed a novel approach called PostDoc that does just that. PostDoc transforms long, multimodal documents (containing text and images) into concise, attractive posters. The challenge lies in selecting the most relevant content while maintaining visual appeal and a coherent narrative. PostDoc tackles this by using a "deep submodular function." This function is trained to identify the most important parts of the document, ensuring good coverage of topics, diversity of information, and alignment between text and images. Once the key content is extracted, PostDoc uses a large language model (LLM) like ChatGPT to rephrase the text into a poster-friendly format. What sets PostDoc apart is its ability to generate a suitable template with design elements like fonts, colors, and layout, all based on the extracted content. The system considers factors like content length and image placement to create a balanced and aesthetically pleasing design. Experiments show that PostDoc outperforms other methods in terms of content quality and design, while also being significantly faster. Human evaluations confirm that PostDoc creates posters that are both informative and visually appealing. While PostDoc currently excels with natural images and text, future improvements will focus on handling more complex visuals like charts and tables, making it an even more powerful tool for researchers and anyone looking to create professional posters effortlessly.