SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Eventual is a Series A AI infrastructure company building Daft, a distributed data engine purpose-built for multimodal AI training. The company is solving a critical bottleneck: today's data platforms lose 20-40% of training time to dataloading alone, while GPU bandwidth grows 2-3× per generation but storage and pipelines haven't kept pace.
As a Systems Engineer on the Dataloading team, you'll design and build the video-native dataloader that transforms multi-petabyte video corpora into GPU-ready tensors at line rate. You'll work with the latest generation hardware (H100, B200, GB200, NVL72) deployed at top AI labs and FAANG companies, optimizing the full data path from object store through NVMe, page cache, host RAM, and device RAM. Your focus: eliminate every avoidable copy and stall, saturate cutting-edge hardware on real customer training jobs, and push toward next-generation bandwidth requirements.
Key responsibilities include designing rank-aware, NVMe-cached dataloaders with random access into clips; profiling and optimizing the complete data path; owning performance benchmarks against customer baselines and historical numbers; partnering with researchers at partner labs to integrate the loader into their training stacks; and collaborating cross-team with Storage Infrastructure and Visual Understanding teams.
You'll bring obsessive attention to systems-level performance, deep OS knowledge (page cache, scheduling, syscalls, NUMA, memory hierarchies), and strong expertise in Rust, C++, or C. You need a precise mental model of where bytes flow: NVMe vs. memory vs. network vs. PCIe vs. NVLink, and their throughput/latency budgets. GPU experience is a plus but not required on day one—the team will uplevel you on CUDA and SLURM. Nice-to-haves include SLURM/Kubernetes experience, video decode pipeline work, and open-source systems contributions.
Eventual has raised $30M from top-tier investors (Felicis, CRV, Microsoft M12, Citi, Y Combinator) and assembled a world-class team from AWS, Render, Pinecone, and Tesla. The team works 4 days/week in-person at the SF Mission office, with competitive compensation, meaningful equity, catered meals, and flexible PTO.
