Deskripsi Pekerjaan
Nexus Intelligence Labs is at the forefront of applied AI research, building foundation models that power the next generation of enterprise automation. We are seeking a Senior Machine Learning Engineer to join our core modeling team in San Francisco. This is a high-impact role where you will own the full lifecycle of large-scale model training, optimization, and deployment to production environments serving millions of API requests daily.
You will work alongside a world-class team of researchers and engineers (ex-Google Brain, Meta AI, OpenAI) in a culture that prioritizes technical excellence, rapid iteration, and intellectual honesty. We offer a competitive equity package, comprehensive healthcare, and a stipend for compute/hardware.
Tanggung Jawab
- Design, implement, and optimize distributed training pipelines for Large Language Models (LLMs) and multimodal architectures on GPU clusters (1000+ GPUs).
- Lead model quantization, distillation, and compilation (TensorRT, Triton, ONNX) for low-latency, high-throughput inference serving.
- Architect robust MLOps infrastructure: feature stores, experiment tracking (MLflow/W&B), CI/CD for model deployment, and automated drift monitoring.
- Collaborate with Research Scientists to translate novel architectures (e.g., Mixture of Experts, State Space Models) from paper to production-grade code.
- Conduct rigorous offline/online evaluation (A/B testing, shadow deployments) to validate model quality improvements.
- Mentor junior engineers and establish best practices for code review, testing, and documentation within the ML platform team.
- Manage cloud compute budgets (AWS/GCP) and optimize resource utilization via spot instances and kernel-level optimizations.
Kualifikasi
- M.S. or Ph.D. in Computer Science, Machine Learning, or equivalent practical experience (5+ years industry experience).
- Expert-level proficiency in Python and Deep Learning frameworks (PyTorch preferred, JAX/TensorFlow a plus).
- Proven track record of training and deploying models at scale (>10B parameters) using distributed strategies (FSDP, DeepSpeed, Megatron-LM).
- Deep understanding of GPU architecture, CUDA kernels, and performance profiling tools (Nsight, PyTorch Profiler).
- Strong software engineering fundamentals: Linux internals, containerization (Docker/K8s), and version control.
- Experience with high-performance data loading and processing (Arrow, Parquet, WebDataset) for petabyte-scale datasets.
- Publications at top-tier venues (NeurIPS, ICML, ICLR, MLSys) or significant open-source contributions to ML infrastructure.