Deskripsi Pekerjaan
About the Role
Nexus AI Labs is seeking a visionary Senior Machine Learning Engineer to architect and deploy next-generation AI systems that power our core products. You will operate at the intersection of research and production, translating complex theoretical concepts into scalable, high-performance solutions used by millions.
We value deep technical expertise, a bias for action, and the ability to mentor engineers across the organization. If you are passionate about pushing the boundaries of LLMs, computer vision, and recommendation systems, we want to hear from you.
What We Offer
- Competitive equity package with quarterly refreshes
- Comprehensive health, dental, and vision coverage (100% employer-paid premiums)
- $3,000 annual learning & development stipend
- Flexible hybrid schedule (3 days onsite in SoMa)
- State-of-the-art GPU compute cluster access
Tanggung Jawab
- Design, train, and optimize large-scale deep learning models (Transformers, Diffusion, GNNs) for production inference.
- Build and maintain robust MLOps pipelines (CI/CD, feature stores, model registry, monitoring) using Kubeflow and MLflow.
- Collaborate with Product and Research teams to define model requirements, evaluation metrics, and data strategies.
- Lead model optimization efforts: quantization, distillation, compilation (TensorRT, ONNX Runtime) for low-latency serving.
- Architect scalable feature engineering platforms handling petabyte-scale structured and unstructured data.
- Conduct rigorous offline/online experimentation (A/B testing, interleaving) to measure business impact.
- Mentor junior engineers, establish coding standards, and drive technical architecture reviews.
- Stay current with SOTA literature (NeurIPS, ICML, ICLR) and prototype novel architectures for internal hackathons.
Kualifikasi
- M.S. or Ph.D. in Computer Science, Statistics, Mathematics, or equivalent practical experience.
- 5+ years of professional experience shipping ML models to production at scale.
- Expert proficiency in Python, PyTorch/TensorFlow/JAX, and distributed training (FSDP, DeepSpeed, Megatron-LM).
- Deep understanding of GPU architecture, CUDA kernels, and memory optimization for inference.
- Proven track record designing ML systems: feature stores, real-time serving, drift detection, automated retraining.
- Strong software engineering fundamentals: Git, Docker, Kubernetes, CI/CD, testing, design patterns.
- Experience with LLMs (fine-tuning, RAG, agents, evaluation frameworks) is highly preferred.
- Excellent communication skills; ability to distill complex technical concepts for cross-functional stakeholders.