SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Mistral is building a sovereign AI cloud for Europe, providing full-stack AI solutions from frontier models to developer tools and compute infrastructure. The Cloud Platform team operates distributed infrastructure across Europe, North America, Asia, and the Middle East, running on proprietary datacenters, networks, and bare metal.
You will design and build the storage layer for Mistral's sovereign AI cloud, supporting training runs, sandboxed workloads, and general compute across European-jurisdiction infrastructure. This is a strategic initiative as Mistral scales toward a gigawatt of compute across Europe by 2030.
In this role, you will:
- Build and own the complete storage stack (block, filesystem, and object storage) for training, sandbox, and general compute workloads
- Design and implement custom services in Go, including S3-compatible object storage APIs, tiering strategies, backup and recovery systems, and caching layers
- Own the data plane: allocation, duplication, tiering, backup and recovery across petabyte-scale deployments
- Plan capacity across datacenters and failure domains at scale
- Partner with Product Managers to shape the storage roadmap and make technical and business tradeoffs
- Work on systems that define how European compute infrastructure stores and manages data
You will decide what storage systems to build internally versus buy from vendors, and harden best-in-class solutions. The systems you build will be mission-critical for customers in finance, healthcare, and the public sector running frontier models on European infrastructure.
REQUIREMENTS:
- 5+ years building backend or infrastructure systems
- Proven experience building storage systems (object, block, or distributed filesystems)—not just operating them. Deep understanding of failure modes (e.g., disk failures mid-write, rebalance collisions with backup windows)
- Strong distributed-systems fundamentals: durability, consistency, replication, erasure coding, failure domains
- Shipped production systems in Go. (Systems engineers from C++/Rust backgrounds ready to write Go daily are encouraged to apply)
- Comfortable with hardware economics: HDD, SSD, NVMe tiers and their cost/performance tradeoffs
- Experience partnering with product, design, and business teams to ship software that reached real users
- Low-ego, team-first mindset; ability to be a technical force other engineers want to work alongside
