About
Altrove is a Paris-based AI startup using machine learning, simulation, and lab automation to reinvent how new materials are discovered. Their platform helps manufacturers identify sustainable alternatives to rare earth elements like cobalt and nickel—critical inputs for EVs, electronics, and defense systems, but often expensive and politically risky to source. After raising €3.7M to bring their platform to market, the Altrove team needed high-performance compute to scale—fast.
Challenge
After using ~$50K in startup credits on AWS to validate their POC, Altrove secured funding and prepared to scale. But costs on hyperscalers were steep—and the H200s they needed weren’t easy to find. Buying their own hardware wasn’t viable either, since their architecture was still evolving. They needed compute that matched their workloads, could be reserved long-term, and didn’t come with quota walls or pricing games.
Solution
Altrove turned to Storj for dedicated H200s that could handle memory-intensive simulations—and were actually available. Storj stood out for its clarity and responsiveness: fast setup, transparent pricing, and sharp technical guidance that helped the team configure a solution without overbuilding. The provider also passed investor diligence, giving Altrove’s backers confidence that the infrastructure would scale with the company.
Result
Storj helped Altrove go from prototype to production with speed and flexibility. They secured the right GPUs, avoided CapEx lock-in, and gained confidence in a long-term partner who could support real-world scale. Altrove continues to expand its compute footprint as the platform grows—building the future of sustainable materials on top of Storj.