Hunyuan Multimodal Algorithm Researcher Intern(Omni-Modal)
Skip the busywork
ApplyBolt rewrites your resume for this exact role and hits submit. You just pick the jobs.
About this role
What the Role Entails
- Conduct research and development of Omni multimodal large models, including the design and construction of training data, foundational model algorithm design, optimization related to pre-training/SFT/RL, model capability evaluation, and exploration of downstream application scenarios.
- Scientifically analyze challenges in R&D, identify bottlenecks in model performance, and devise solutions based on first principles to accelerate model development and iteration, ensuring competitiveness and leading-edge performance.
- Explore diverse paradigms for achieving Omni-modal understanding and generation capabilities, research next-generation model architectures, and push the boundaries of multimodal models.
Who We Look For
1. Bachelor’s degree (full-time preferred) or higher in Computer Science, Artificial Intelligence, Mathematics, or related fields; graduate degrees are prioritized.
2. Hands-on experience in large-scale multimodal data processing and high-quality data generation is highly preferred.
3. Solid foundation in deep learning algorithms and practical experience in large model development; familiarity with Diffusion Models and Autoregressive Models is advantageous. Publication in top-tier conferences or experience in cross-modal (e.g., audio-visual) research is preferred.
4. Proficiency in underlying implementation details of deep learning networks and operators, model tuning for training/inference, CPU/GPU acceleration, and distributed training/inference optimization; practical experience is a plus.
5. Participation in ACM or NOI competitions is highly valued.
6. Strong learning agility, communication skills, teamwork, and curiosity.