This job has expired

This position was posted on October 24, 2025 and is likely no longer accepting applications. We've kept it here for historical reference. Check out the similar jobs below!

Principal Engineer, CPU Design, AI ML Applications

Tenstorrent Inc.

Job Overview

Location

Remote

Job Type

Full-time

Full Job Description

📋 Description

• Own the end-to-end architecture and micro-architecture of Tenstorrent’s next-generation RISC-V CPU cores that will power the world’s most demanding AI/ML workloads. You will translate high-level performance targets into detailed RTL specifications, balancing power, area, and frequency trade-offs while pushing the boundaries of what is possible in AI acceleration.
• Drive the definition of new instruction-set extensions, memory-hierarchy features, and specialized functional units that accelerate tensor operations, sparse-matrix math, and transformer kernels. Your designs will directly influence compiler back-ends, runtime libraries, and model-deployment pipelines used by thousands of developers.
• Lead cross-functional design reviews with software, verification, physical design, and silicon-validation teams to ensure that every architectural decision is validated early and often. You will champion a “shift-left” culture where performance, power, and security are analyzed from day one, not after tape-out.
• Mentor senior and junior engineers across geographies, establishing coding guidelines, micro-architectural templates, and reusable IP blocks that scale from edge devices to multi-die data-center chiplets. Your leadership will shape the technical DNA of Tenstorrent’s CPU roadmap for the next decade.
• Prototype critical features in high-level models (C++/SystemC/TLM) and then refine them in synthesizable RTL (Chisel/Verilog/SV). You will use performance simulators, FPGA rigs, and silicon bring-up labs to measure real-world speed-ups on production AI graphs, iterating until every cycle counts.
• Collaborate with AI research teams to understand emerging model architectures—from CNNs to diffusion and large-language models—and translate their compute patterns into CPU enhancements. You will publish white-papers and speak at conferences, positioning Tenstorrent as a thought leader in AI-centric CPU design.
• Establish rigorous power-performance-area (PPA) dashboards that track every micro-architectural knob. You will run design-space explorations with EDA tools and machine-learning optimizers to squeeze out every milliwatt and square millimeter while hitting aggressive GHz targets.
• Champion security-by-design, integrating features such as memory tagging, side-channel resistance, and confidential-compute primitives into the CPU micro-architecture. You will work with OS and firmware teams to ensure these features are exposed through open standards and upstreamed to the RISC-V community.
• Influence the company’s long-term silicon roadmap, from 3 nm process choices to 2.5D/3D packaging strategies. You will partner with foundry and OSAT vendors to co-optimize die floor-planning, thermal design, and high-bandwidth interconnects for exascale AI clusters.
• Foster an inclusive, remote-first culture where curiosity and respectful debate are celebrated. You will run weekly architecture deep-dives, brown-bag sessions, and hackathons that keep the team at the cutting edge of CPU and AI research.

🎯 Requirements

• 10+ years of hands-on CPU micro-architecture and RTL design experience, including at least one tape-out in 7 nm or smaller nodes
• Expert-level knowledge of RISC-V or other modern ISA extensions, pipeline design, cache hierarchies, and vector/SIMD units
• Proven track record of delivering >1 GHz high-performance cores with best-in-class PPA metrics
• Proficiency with hardware description languages (Chisel, SystemVerilog) and performance modeling frameworks (gem5, Sniper, or proprietary)
• Nice-to-have: published papers or patents in AI/ML acceleration, security, or advanced packaging technologies

🏖️ Benefits

• Fully remote-first culture with flexible hours and quarterly in-person innovation summits in global tech hubs
• Competitive salary, equity, and performance bonuses tied to product milestones and silicon success
• Annual learning stipend of $5,000 for conferences, courses, and research publications
• Comprehensive health, dental, and vision coverage for you and dependents, plus a monthly wellness allowance
• Home-office setup budget and top-tier compute resources (cloud and on-prem FPGA farms) for experimentation

Skills & Technologies

Senior

Remote

Ready to Apply?

Apply Externally

You will be redirected to an external site to apply.

Tenstorrent Inc.

Visit Website

About Tenstorrent Inc.

Tenstorrent designs high-performance RISC-V and AI processors and licensable IP for data-center and edge systems. Founded by CPU architect Ljubisa Bajic and chip legend Jim Keller, the company combines scalable Tensix cores with open-source software to accelerate training and inference workloads. Its products include Grayskull, Wormhole and Blackhole PCIe cards, plus the Ascalon RISC-V core and DevCloud services. Headquartered in Toronto with offices in Austin, Santa Clara and Belgrade, Tenstorrent serves hyperscalers, OEMs and research institutions seeking efficient, programmable compute for AI, HPC and automotive applications.

View Company Profile