Member of Technical Staff, Multimedia (Vision)

The Role: 

We are looking for a highly motivated Member of Technical Staff with expertise in vision-language modeling to join our research and engineering team. This role will drive advancements in our multimodal models and applications that combine visual understanding with natural language. You’ll be responsible for conducting cutting-edge research and building production-quality systems that bring state-of-the-art VLM capabilities into real-world products.

Key Responsibilities: 

• Lead research and development efforts in vision-language models, including data preparation, model training, evaluation, and deployment.
• Collaborate with teams in engineering, product, and design, and work directly with customers to understand their needs and translate VLM capabilities into real-world applications.
• Design and implement scalable machine learning systems for tasks such as image captioning, visual question answering, retrieval, grounding, and multimodal reasoning.
• Train large-scale VLMs using techniques such as  parameter-efficient fine-tuning (LoRA/QLoRA), reinforcement learning approaches, dataset curation and preparation, distributed training (DDP/FSDP), hyperparameter optimization.
• Build robust, maintainable code in Python for both experimentation and production use.
• Analyze model performance, conduct rigorous evaluations, and experiment based on empirical insights.
• Contribute to the platform roadmap by providing technical insights into quality improvements, integrating latest multi-modal research, and identifying and proposing new platform capabilities with significant commercial potential.

Minimum Qualifications: 

• Bachelor’s degree in Computer Science, Electrical Engineering, or a related field.
• 3 years of experience in machine learning, with a focus on computer vision, NLP, or multimodal systems.
• Strong proficiency in Python and deep learning frameworks such as PyTorch or TensorFlow.
• Experience training and deploying large-scale models and working with distributed computing environments.
• Demonstrated ability to write production-quality code and collaborate across teams.
• Experience working directly with customers, partners, or external stakeholders to define use cases or requirements.

Preferred Qualifications: 

• Master’s or PhD in a relevant technical field with research experience in vision-language or multimodal modeling.
• Experience with multimodal training/fine-tuning and downstream tasks like VQA, captioning, or retrieval.
• Familiarity with large language models (LLMs) and their integration with visual encoders.
• Contributions to open-source projects or publications in top-tier ML/AI conferences (e.g., CVPR, ICCV, NeurIPS, ICML, ACL).
• Comfortable working in fast-paced, cross-disciplinary environments and shipping research into production.