Our Client:
Fintech Open Platform – Dedicated to driving the digital upgrade of global modern services, including the financial services industry, through technology.
Responsibilities
- Innovate Large Model Applications: Develop innovative solutions to enhance controllable knowledge production and inference efficiency in large models.
- Solve Technical Challenges: Address complex decision-making capabilities of agents to remove barriers in large model implementations.
- Collaborate Across Teams: Partner with technical and business stakeholders to align efforts and achieve common goals effectively.
- Explore NLP Innovations: Investigate cutting-edge NLP large models to drive advancements and foster innovation within subfields.
- Establish Technical Leadership: Build and maintain industry-leading technical capabilities in NLP through rigorous research and development efforts.
Requirement:
- Bachelor’s degree in Computer Science or a related field; a Master is preferred. Prior publication of research papers in top AI conferences related to large models is a plus.
- Strong foundational background in NLP, with a deep understanding of mainstream large models such as GPT-3, ChatGPT, T5, PaLM, LLaMA, GLM, etc., and practical project experience is preferred.
- Proficient in mainstream deep learning frameworks such as PyTorch/TensorFlow, and experience with large model training frameworks like Megatron-LM/DeepSpeed for multi-machine, multi-GPU solutions. Experience in training and fine-tuning large NLP models with billions or hundreds of billions of parameters is preferred.
- Familiar with common model compression techniques such as quantization, pruning, and distillation, and knowledgeable in ONNX/TensorRT.
- Strong coding skills, with experience in open-source project development preferred. Good communication skills and project leadership experience are essential.