6 months ago
Base Salary
$140k - $180k/yr
Responsibilities
- Build backend and hardware architecture for high-quality, high-speed research.
- Own the data lifecycle across pipelines processing video, audio, text, and time-series datasets.
- Provision and manage cloud and bare-metal compute clusters used for model training.
- Connect physical neurotechnology hardware to central repositories and provide researchers with high-throughput, low-latency access to GPUs.
- Define storage topologies, manage databases, and synchronize datasets between on-premises edge servers and cloud environments.
- Optimize low-level I/O-bound operations and own infrastructure through production deployment.
Requirements
- At least 3 years of production software engineering experience with deep systems-level architecture expertise.
- Expertise in one or more of Python, Rust, C++, or Go.
- Experience building and maintaining high-performance ETL pipelines processing, buffering, and storing terabytes of daily unstructured data.
- Experience architecting, provisioning, and maintaining bare-metal compute clusters, storage servers, and high-speed networking for ML workloads.
- Experience handling continuous, highly concurrent data streams from heterogeneous hardware peripherals without data loss.
- Ability to define storage topologies, manage databases, and synchronize massive datasets across on-premises edge servers and the cloud.
- Preferred experience with PyTorch or TensorFlow and building datasets that maximize GPU utilization.
- Preferred experience with InfiniBand, RoCE, zero-copy networking, or shared memory for distributed GPU training.
- Preferred experience with FFmpeg, GStreamer, or OpenCV for programmatic video processing.
Benefits
- Competitive equity compensation at a seed-stage startup
- Housing support options
- Visa sponsorship
- 3% 401(k) matching
- Health insurance
About Alljoined
Alljoined aims to decode thoughts from the brain, entirely non-invasively. We apply deep learning research to large-scale EEG datasets to decode images, emotion, intent, train of thought, and more. Our goal is to create a general interface that enables people to better understand themselves, communicate more effectively, and connect more naturally with technology. We are actively growing a world-class team of researchers to build the most performant and accessible interface possible, with the aim of meaningfully improving the quality of individual lives as well as the well-being of society as a whole.