Eventual

You'll contribute to **Daft's distributed query engine** and **video-native dataloading layer** for Physical AI systems. Your work will move petabytes of video through GPU clusters at top research labs. You'll ship code that runs on the most expensive GPU clusters on the planet.

You'll build the **video-native dataloader** for a **distributed data engine** powering Physical AI training. Your work keeps **B200 and NVL72 GPUs** fully saturated. You'll partner with top AI labs and own performance from NVMe to GPU memory.

You'll own the layer that makes petabytes of video queryable by content, defining the roadmap for **visual understanding capabilities** and shipping to production. Your work directly impacts customer training runs, turning raw footage into **curated datasets in minutes**.

You'll own the product surface for Physical AI researchers to explore, query, and curate **multimodal datasets** at **Eventual**, a startup building a video-native index. Your work directly enables faster AI model iteration, and you'll collaborate closely with researchers at top labs.