AI Engineer Intern – Foundation Models for Financial Applications (Hong Kong)

Location

Hong Kong

Key Responsibilities

• Assist in training and fine-tuning large-scale foundation models for financial applications
• Support scaling of training runs across multi-GPU/TPU clusters
• Contribute to building and optimizing inference pipelines for deployment
• Help implement and experiment with distributed training strategies
• Develop and test custom JAX/PyTorch implementations for novel architectures
• Assist in monitoring and debugging training runs using experiment tracking tools

Essential Requirements

• Exposure to training or fine-tuning deep learning models (experience with 100M+ parameter models is a plus)
• Familiarity with JAX and/or PyTorch
• Basic understanding of multi-GPU training concepts
• Familiarity with transformer architectures and attention mechanisms
• Exposure to distributed training frameworks (DeepSpeed, FSDP, Accelerate, or similar)
• Familiarity with foundation model fine-tuning techniques (LoRA, QLoRA, PEFT)

Nice to Have

• Background in High Frequency Trading (HFT) or financial applications
• Exposure to diffusion models or autoregressive architectures
• Familiarity with cloud platforms for ML workloads (GCP, AWS, or similar)
• Familiarity with MLOps tools (Weights & Biases, MLflow)
• Any coursework or projects touching on mixture-of-experts, quantization, or model compression

What We Offer

• Hands-on experience working on cutting-edge AI for major financial institutions (LSEG, Barclays)
• Access to significant compute resources (GPUs/TPUs)
• Competitive internship stipend
• Mentorship from senior AI engineers with direct impact on core technology
• Potential for full-time conversion upon successful completion