Machine Learning Engineer
NodeFlair
- Singapore
- Permanent
- Full-time
- Guarantee the stability of inference and training processes across various hardware and software configurations, including:
- CPU Hardware:
- Popular consumer and server-grade processors from major vendors (Intel, AMD, Apple)
- GPU/AI Accelerators:
- Leading solutions from NVIDIA (CUDA, Triton Inference Server), AMD (ROCm), Intel (Arc, NPU), and Apple (Metal/CoreML)
- Model Formats:
- Industry-standard formats like GGUF and PyTorch
- Operating Systems:
- Windows, macOS, and Linux
- Inference Engines:
- Widely used engines like llama.cpp, vLLM, and TensorRT-LLM
- Training Frameworks:
- Popular frameworks like PyTorch and MLX
- (Bonus) Expertise in edge training/fine-tuning, micro training, and federated training across diverse hardware and software AI stacks.
- Solid software engineering skills with proven experience in Python and C++
- Effective communication skills in English
- Familiarity with essential development tools and workflows like Git, CI/CD, and Agile methodologies
- Experience in hardware optimization for CPUs, GPUs, and NPUs
- Exposure to large language models (LLMs), including popular foundation models like Llama2, Mistral, and GPT-2
- (Bonus) Experience deploying LLMs/AI models in production environments (server/edge)
- Competitive all-inclusive compensation package
- Generous leave policy (14 days + unlimited sick leave)
- Annual equipment budget available after probation period