Tenstorrent

Build and improve reliable, high-performance infrastructure for silicon development, including simulation, verification, EDA, and CI/CD workflows. The role combines distributed systems, operational reliability, developer productivity, and AI-assisted engineering in a hybrid Austin environment.

Build compiler and runtime workflows that help customers integrate AI models with Tenstorrent’s AI acceleration IP. This hands-on software role spans model compatibility, graph transformations, runtimes, and customer-facing technical problem solving.

Lead the design and implementation of low-level runtime software that controls Tenstorrent’s custom processors, accelerators, memory systems, and data movement. This hybrid Toronto/Austin role combines hands-on C/C++ development, hardware bring-up, performance optimization, and technical leadership.

Optimize real-world datacenter and agentic AI workloads across runtimes, compilers, software systems, and RISC-V CPU architecture. This remote role uses profiling, benchmarking, and hardware-aware optimization to shape both the RISC-V software ecosystem and future CPU designs.

Build low-level kernels and software primitives that directly program Tenstorrent’s custom AI hardware. This hybrid Toronto role focuses on hardware-aware optimization, performance engineering, and enabling efficient AI workload execution.

Build production-ready system-level test software and infrastructure that validates AI silicon under real workloads and scales into high-volume manufacturing. You’ll lead cross-domain debugging across hardware, firmware, operating systems, and test systems to improve yield and product quality.

Own the technical outcomes of customer AI deployments, integrating hardware and software and driving systems through validation, production readiness, and ongoing operations. This hybrid Tokyo-based role combines hands-on engineering, customer collaboration, debugging, and deployment ownership.

Build and operate performance infrastructure for RISC-V software and performance teams, automating benchmarking, data collection, workload capture, and analysis workflows. This hybrid role spans silicon, FPGA/emulation platforms, performance models, and CPU/AI systems.