Datacenter & Agentic AI Workload Performance Optimization Engineer
Tenstorrent · Santa Clara, California, United States; Toronto, Ontario, Canada · North America
CHECKED 22H AGO
Infrastructureonsite$100k–$500k
At a glance
Highlights
- Optimize datacenter and agentic AI workloads to improve throughput and efficiency.
- Influence future CPU architecture by linking real workload behavior to design.
- Build AI-assisted workflows that continuously profile and optimize software.
Ideal candidate
A curious performance engineer who loves digging into why software is fast or slow and collaborating across hardware and software teams. You care about efficiency at scale and using AI tools to accelerate optimization.
Requirements
- Master's or PhD in Computer Engineering, Electrical Engineering, Computer Science or related field
- Experience in performance optimization, computer architecture, compilers or systems software
- Hands-on experience with runtime or compiler optimization (e.g. OpenJDK/JIT, LLVM, GCC, V8, Python)
- Understanding of CPU performance, memory hierarchies, concurrency, garbage collection and vector/SIMD
- Expertise with profiling tools such as Linux perf, runtime profilers, QEMU and tracing tools
- Strong programming skills in Java, Python, C/C++ and RISC-V assembly across software layers
- remote policy
- Remote, based in North America
- hiring process
- Candidates will be assessed for appropriate level during interview process, offers align with that level
Lab profile
About Tenstorrent
Builds high-performance RISC-V CPUs and AI accelerators with full-stack software for efficient AI computing.
What they build
- AI accelerators
- RISC-V CPUs
- AI software and compilers
Known for
- Custom RISC-V CPU development from scratch
- Full-stack AI hardware and software
- Performance, ease of use and cost efficiency
Key facts
- Founded in 2016
- Headquartered in Toronto, Canada
- Backed by Samsung and Hyundai Motor Group