Skip to content
Search jobs inside your AI agent. Claude, ChatGPT, Claude Code, Codex, Gemini, Cursor and Grok
Connect

Tenstorrent

NewBe an early applicant

Staff Engineer, OPs

Posted Updated

Description

You will design, implement and optimize high-performance, numerically accurate operations for the latest LLMs and text, image and video models on Tenstorrent hardware. You will debug precision issues, build tests and tooling, design agentic workflows for generating and tuning operations, and deliver measurable performance gains through benchmarking and profiling, collaborating on software/hardware co-design.

Responsibilities

  • Design, implement, and optimize high-performance, numerically accurate operations for LLMs and text, image, and video models
  • Debug precision issues and build tests and tooling to catch them early
  • Design agentic workflows to generate, test, and tune operations
  • Define metrics and processes for evaluating agentic code generation
  • Deliver performance improvements through benchmarking and profiling
  • Collaborate with models, runtime, and hardware teams on software/hardware co-design

Requirements

  • Strong background in low-level software development in C++ or C
  • Focus on performance optimization and understanding of CPU, GPU, or accelerator architecture
  • Comfort with numerical analysis, floating-point arithmetic, and accuracy across data formats
  • Experience using agentic AI to automate software development
  • Eligibility to access U.S. export-controlled technology

Skills

Apply

See also