2 hours ago
Responsibilities
- Build multimodal storage systems using modern columnar data lake formats for video, lidar, and sensor logs.
- Develop query engine capabilities including partitioning, indexing, query planning, embeddings/vector search, and perception-based query predicates.
- Improve dataloading memory stability, throughput, and zero-copy streaming to deliver CUDA tensors to GPUs at line rate.
- Work across compute infrastructure, data storage and querying layers, and model training and deployment systems.
Requirements
- Proven experience building resilient, high-throughput distributed systems or database engines using Rust or C++.
- At least 3 years of experience working deeply on engine internals such as vectorized execution, query planning and optimization, distributed task scheduling, or zero-copy networking.
- Practical experience scaling cloud infrastructure, particularly AWS S3.
- Experience with CUDA, GPU streaming, or video decoding frameworks is preferred.
- Ability to work autonomously and adapt in a fast-paced startup environment.
Benefits
- In-person role with four days per week in the San Francisco office.
- Competitive compensation and startup equity.
- Catered lunches and dinners for San Francisco employees.
- Commuter benefit.
- Team-building events and poker nights.
- Health, vision, and dental coverage.
- Flexible paid time off.
- Latest Apple equipment.
- 401(k) plan with employer match.
About Eventual
Eventual is building a multimodal data platform for AI systems from the ground up, designed to tackle the challenges of working with traditional data engineering and analytics alongside modern ML/AI workloads. Eventual has raised $30M from investors including Felicis, CRV, M12, Citi, YCombinator, Array VC, Caffeinated Capital and top Silicon Valley executives and founders in companies such as Meta, Lyft and Databricks.