Block is looking for an experienced Machine Learning Engineer to join their ML Inference & Training team to build and improve ML infrastructure, platforms, and libraries that enable applied ML teams across Block to develop, productionize, and observe ML models at scale.
Requirements
- Deep understanding of architectures for ML-backed solutions and the ML lifecycle - from data gathering, training, model evaluation, MLOps, and productionizing models (both classical ML and deep learning/LLMs)
- Ability to produce production-quality code and services incorporating testing, evaluation, monitoring as well as the ability to quickly adapt to a new domain, hack MVP's, and iterate to improve product
- Experience using any of the major cloud vendors for high scale production use cases
- Golang, Python, TypeScript, React
- gRPC, Protocol Buffers
- AWS (Sagemaker, EC2, EKS), Databricks
- NVIDIA Triton, vLLM, ONNX
Responsibilities
- Develop scalable ML/AI systems and platforms
- Be on the cutting edge of advancements in ML/AI across the industry
- Collaborate across diverse teams to deliver impactful platform solutions by understanding and generalizing the problems of cross-functional stakeholders
- Support critical ML systems used for fraud detection, language models, recommendation systems and underwriting
- Influence our roadmap and ensure we're working on the highest impact problems within the ML infrastructure domain
Other
- 8+ years of experience in software development and/or machine learning experience with a focus on internal platforms or infrastructure
- A desire to understand our clients' needs in order to design tools and systems that solve our customer's problems. As a platform team, our main customers are internal developers and products
- Strong communication skills (verbal and written) with technical and non-technical stakeholders
- We value general and versatile engineering skill over specific experience with our current stack
- Block is an equal opportunity employer evaluating all employees and job applicants without regard to identity or any legally protected class.
