Job
Description
As a Vision-Language Model Developer, you will be responsible for developing, fine-tuning, and evaluating vision-language models such as CLIP, Flamingo, BLIP, GPT-4V, LLaVA, etc. You will design and build multimodal pipelines that integrate image/video input with natural language understanding or generation. Working with large-scale image-text datasets like LAION, COCO, and Visual Genome for training and validation will be a key part of your role. Implementing zero-shot/few-shot multimodal inference, retrieval, captioning, VQA (Visual Question Answering), grounding, etc., will also be within your responsibilities. Collaboration with product teams, ML engineers, and data scientists to deliver real-world multimodal applications is essential. Additionally, optimizing model inference performance and resource utilization in production environments using ONNX, TensorRT, etc., will be part of your duties. You will also conduct error analysis, ablation studies, and propose improvements in visual-language alignment. Contribution to research papers, documentation, or patents, if in a research-driven team, is expected. Qualifications required for this role include a Bachelors/Masters/PhD in Computer Science, AI, Machine Learning, or a related field. You should have at least 2+ years of experience in computer vision or NLP, with a minimum of 1+ year in multimodal ML or VLMs. Strong programming skills in Python, with experience in libraries like PyTorch, HuggingFace Transformers, OpenCV, and torchvision are necessary. Familiarity with VLM architectures such as CLIP, BLIP, Flamingo, LLaVA, Kosmos, GPT-4V, etc., is expected. Experience with dataset curation, image-caption pair processing, and image-text embedding strategies is also required. A solid understanding of transformers, cross-attention mechanisms, and contrastive learning is essential for this role. Please note that this is a full-time position with a day shift schedule. The work location is in person. As a Vision-Language Model Developer, you will be responsible for developing, fine-tuning, and evaluating vision-language models such as CLIP, Flamingo, BLIP, GPT-4V, LLaVA, etc. You will design and build multimodal pipelines that integrate image/video input with natural language understanding or generation. Working with large-scale image-text datasets like LAION, COCO, and Visual Genome for training and validation will be a key part of your role. Implementing zero-shot/few-shot multimodal inference, retrieval, captioning, VQA (Visual Question Answering), grounding, etc., will also be within your responsibilities. Collaboration with product teams, ML engineers, and data scientists to deliver real-world multimodal applications is essential. Additionally, optimizing model inference performance and resource utilization in production environments using ONNX, TensorRT, etc., will be part of your duties. You will also conduct error analysis, ablation studies, and propose improvements in visual-language alignment. Contribution to research papers, documentation, or patents, if in a research-driven team, is expected. Qualifications required for this role include a Bachelors/Masters/PhD in Computer Science, AI, Machine Learning, or a related field. You should have at least 2+ years of experience in computer vision or NLP, with a minimum of 1+ year in multimodal ML or VLMs. Strong programming skills in Python, with experience in libraries like PyTorch, HuggingFace Transformers, OpenCV, and torchvision are necessary. Familiarity with VLM architectures such as CLIP, BLIP, Flamingo, LLaVA, Kosmos, GPT-4V, etc., is expected. Experience with dataset curation, image-caption pair processing, and image-text embedding strategies is also required. A solid understanding of transformers, cross-attention mechanisms, and contrastive learning is essential for this role. Please note that this is a full-time position with a day shift schedule. The work location is in person.
