Datacenter & Agentic AI Workload Performance Optimization Engineer

Job Title

Datacenter & Agentic AI Workload Performance Optimization Engineer

Role Summary

Work on performance optimization for datacenter and agentic AI software stacks running on Tenstorrent RISC-V platforms. The role spans runtimes, compilers, systems software, and CPU microarchitecture to improve throughput, latency, and efficiency for real workloads.

You'll profile applications, develop compiler/runtime/hardware-aware optimizations, use RISC-V Vector/Matrix and custom ISA features, and help automate AI-assisted optimization workflows. This is a remote role based in North America.

Experience Level

Mid-level. The posting requests advanced technical experience in performance optimization and systems/software-hardware co-design; no explicit years-of-experience range was provided.

Responsibilities

Primary responsibilities include profiling, diagnosing, and improving performance across software layers for datacenter and AI workloads.

• Profile real-world applications and runtimes to identify bottlenecks (latency, throughput, memory, concurrency).
• Develop and implement compiler, runtime, and software optimizations informed by CPU microarchitecture and RISC-V capabilities.
• Bring up, tune, and benchmark major runtimes and language ecosystems (Java, Python, Node.js, etc.).
• Explore and prototype use of RISC-V Vector, Matrix, and custom instructions to accelerate workloads.
• Use and extend profiling, tracing, simulation, and performance-modeling tools; iterate on hypotheses with data-driven changes.
• Create and maintain automated and AI-assisted workflows for profiling, code generation, and benchmarking.
• Collaborate across compiler, runtime, systems, hardware, and architecture teams to translate workload behavior into design or software changes.

Requirements

Must-have technical skills and experience; education requirements are summarized separately below.

• Hands-on experience optimizing software on RISC-V or another modern CPU architecture.
• Practical experience with runtime or compiler optimization (examples: OpenJDK/JIT, LLVM, GCC, V8, Python runtime internals).
• Solid understanding of CPU performance, memory hierarchies, concurrency, garbage collection, and vector/SIMD optimization.
• Experience with profiling and analysis tools such as Linux perf, runtime profilers, QEMU, tracing tools, and performance-modeling environments.
• Strong programming skills in Java, Python, and C/C++; familiarity with RISC-V assembly.
• Ability to form data-driven hypotheses, iterate on optimizations, and communicate results to cross-functional teams.

Nice-to-have:

• Experience with AI-assisted development or automated optimization workflows.
• Exposure to datacenter-scale workload tuning and performance modeling for production systems.

Education Requirements

Master's or PhD in Computer Engineering, Electrical Engineering, Computer Science, or a related field (as stated in the posting).

About the Company

Company: Tenstorrent

Headquarters: Austin, Texas, United States

Tenstorrent is a technology company focused on designing innovative computing solutions. They are known for their expertise in the development of advanced hardware, including ASICs and SoCs, aimed at enhancing performance and efficiency in various applications.