HPC for Quant Trading Firms running backtesting at scale.
Distributed Compute for AI
Quant research is generating candidate strategies orders of magnitude faster than backtesting infrastructure can evaluate them. AI has further accelerated alpha research from a handful of signals per quarter to hundreds per day. The bottleneck has moved from researcher time to compute throughput. We're building large HPC clusters (100 - 50,000 nodes) that can run up to 6.4M tasks in parallel across multiple cloud providers on spot instances. Our goal is to make backtesting strategies simple, fast, and cheap at scale.
We build distributed compute clusters with the cheapest CPUs and GPUs across Hyperscalers and Neoclouds for AI. Our mission is to bring frontier-grade infrastructure to everyone. We believe the bottleneck isn't just GPU availability, but effective utilization by the infrastructure software that runs on it. We're starting by building large scale high performance computing (HPC) clusters for quantitative trading firms to run parallel simulation workloads such as backtesting. Our technology generalizes to critical AI workloads such as post-training with reinforcement learning, fine-tuning, long-horizon agents with high tool use, and batch inference.
Expanded from HPC specifically for quant trading backtesting to distributed compute for broader AI workloads including training and fine-tuning, representing a significant market expansion while leveraging the same core distributed computing technology.