Senior Research Engineer: Efficient Generative AI Models

TikTok; San Jose, Northern (CA, KY); Híbrido; USD 254 000 - 588 000; Tempo integral

Vantagens oferecidas por esta oferta de emprego

Medical/dental/vision insurance

401(k) match

Paid parental leave

Disability insurance

Life insurance

Wellbeing benefits

Holidays and PTO

Resumo da oferta

TikTok's Vision-Applied Research team in San Jose is seeking a Sr. Research Engineer/Scientist to design and implement efficient models for large-scale generative AI, focusing on distillation and compression for scalable deployment.

The role involves developing methods and infrastructure for transferring capabilities from foundation models to smaller, hardware-friendly models, including distillation, quantization, and model acceleration, with cross-functional collaboration.

Qualificações

• B.S. in Computer Science or related fields, or equivalent experience.
• Expertise in efficient models with understanding of bottlenecks and acceleration.
• Experience with training generative AI or LLM models using PyTorch and JAX.
• Strong communication and collaboration in fast-paced environments.

Responsabilidades

• Develop efficient algorithms and architectures for large-scale generative and multimodal models, using techniques such as distillation and quantization to improve efficiency.
• Advance scalable generative modeling approaches, including diffusion and autoregressive models, with a focus on acceleration and efficiency.

Conhecimentos

Efficient model optimization

Strong communication

Collaborative work

GenAI expertise

Formação académica

B.S. in Computer Science or related fields

Ph.D. in GenAI or MLSys

Ferramentas

PyTorch

JAX

Descrição da oferta de emprego

TikTok's Vision-Applied Research team in San Jose is seeking a Sr. Research Engineer/Scientist to design and implement efficient models for large-scale generative AI, focusing on distillation and compression for scalable deployment.

The role involves developing methods and infrastructure for transferring capabilities from foundation models to smaller, hardware-friendly models, including distillation, quantization, and model acceleration, with cross-functional collaboration.