Datacenter & Agentic AI Workload Performance Optimization Engineer

Job summary

Responsibilities

Compensation for all engineers at Tenstorrent ranges from $100k - $500k including base and variable compensation targets. Experience, skills, education, background and location all impact the actual offer made.

• How to optimize modern software stacks from application and runtime all the way down to CPU microarchitecture and silicon.
• How RISC-V Vector, Matrix, and custom ISA capabilities can be used to accelerate real-world datacenter and AI workloads.
• How runtime, compiler, memory, and concurrency decisions impact performance at datacenter scale.
• How to build automated and AI-assisted performance optimization workflows that continuously profile, analyze, modify, and benchmark software.
• How to influence future CPU architecture by connecting real workload behavior and software optimization opportunities to hardware design decisions.

What they're looking for

• You’re a performance engineer who enjoys getting deep into runtimes, compilers, applications, and CPU microarchitecture to understand why software is fast—or slow.
• You have hands-on experience optimizing software on RISC-V or another modern CPU architecture, with a strong understanding of the hardware/software boundary.
• You’re comfortable profiling complex systems, finding bottlenecks, forming hypotheses, and iterating through optimizations using data.
• You’re excited about emerging agentic AI development workflows and using AI tools to automate profiling, coding, benchmarking, and optimization.
• You’re a strong technical collaborator who can work across compiler, runtime, systems software, hardware, and performance modeling teams.

Tech stack

• Java
• Python
• Node.js
• Go
• Rust
• C++