cover image
Tenstorrent

Tenstorrent

www.tenstorrent.com

3 Jobs

887 Employees

About the Company

Tenstorrent is a next-generation computing company that builds computers for AI.

Headquartered in the U.S. with offices in Austin, Texas, and Silicon Valley, and global offices in Toronto, Belgrade, Seoul, Tokyo, and Bangalore, Tenstorrent brings together experts in the field of computer architecture, ASIC design, RISC-V technology, advanced systems, and neural network compilers. Tenstorrent is backed by Eclipse Ventures and Real Ventures, Archerman Capital, Samsung Catalyst Fund, and Hyundai Motor Group among others.

Join us: www.tenstorrent.com/careers.

Listed Jobs

Company background Company brand
Company Name
Tenstorrent
Job Title
Sr. Software Engineer, AI Compiler
Job Description
**Job Title:** Sr. Software Engineer, AI Compiler **Role Summary:** Design and develop TT‑Forge, an open‑source MLIR‑based compiler that enables AI workloads on Tenstorrent hardware. Collaborate with hardware and software teams to create and optimize custom compiler dialects, transformation passes, and human‑in‑the‑loop tuning tools, driving performance and scalability for diverse AI models. **Expactations:** - Deliver high‑quality, production‑ready compiler components. - Solve complex AI compilation challenges from graph optimization to multi‑chip scaling. - Contribute to open‑source projects and foster cross‑functional collaboration. - Continuously learn emerging compiler and AI technologies. **Key Responsibilities:** - Implement and maintain custom MLIR dialects (TTIR, TTNN, TTKernel) and associated passes. - Optimize computational graphs for RISC‑V AI accelerators. - Develop tooling (e.g., TT‑Explorer) for interactive compiler tuning. - Integrate with high‑level frameworks such as PyTorch, JAX, and TensorFlow. - Conduct performance analysis and benchmark AI workloads across hardware configurations. - Participate in code reviews, design discussions, and documentation of compiler architecture. **Required Skills:** - Strong proficiency in C++ (modern standards) and Python. - Experience with compiler construction, preferably using MLIR or LLVM. - Familiarity with AI/ML frameworks (PyTorch, JAX, TensorFlow). - Knowledge of low‑level hardware interaction and RISC‑V architecture a plus. - Ability to design, implement, and debug complex systems. - Effective communication and teamwork in an open‑source, collaborative environment. **Required Education & Certifications:** - Bachelor’s degree in Computer Science, Electrical Engineering, or related field (Master’s preferred). - Relevant coursework or project experience in compilers, systems programming, or AI hardware. - No specific certifications required; demonstrated technical expertise is essential.
Toronto, Canada
On site
03-03-2026
Company background Company brand
Company Name
Tenstorrent
Job Title
Sr. Engineer, Software - AI Compiler
Job Description
Job Title: Sr. Engineer, Software – AI Compiler Role Summary: Lead design, implementation, and optimization of TT-Forge, an MLIR‑based compiler enabling AI workloads on Tenstorrent hardware. Drive development of custom MLIR dialects, transformation passes, and human‑in‑the‑loop tuning tools to support training, inference, and multi‑chip scaling across diverse AI frameworks. Collaborate closely with hardware, firmware, and system teams to integrate low‑level execution backends and expose high‑performance execution paths to developers. Expectations: - Deliver production‑ready compiler components that bridge popular ML frameworks (PyTorch, TensorFlow, JAX) to RISC‑V‑based AI chips. - Reduce latency, memory footprint, and power consumption through advanced graph optimizations and hardware‑aware scheduling. - Extend MLIR infrastructure to include custom dialects (TTIR, TTNN, TTKernel) and maintain maintainable source code with comprehensive tests and documentation. - Engage with open‑source community contributions and incorporate external tooling for profiling, debugging, and fine‑grained tuning. - Mentor junior engineers and provide architectural guidance within the compiler group. Key Responsibilities: 1. Design and implement MLIR dialects, passes, and optimization pipelines for AI model translation. 2. Build and maintain converters from high‑level frameworks to TT-Forge IR, ensuring correct semantics and optimal performance. 3. Develop compiler backends targeting multi‑chip TT hardware, managing data movement, pipelining, and resource allocation. 4. Create and refine tools for profiling, debugging, and human‑in‑the‑loop tuning (e.g., TT‑Explorer). 5. Collaborate with hardware engineers to expose new ISA features and verify compiler outputs against hardware simulations. 6. Maintain high code quality through automated testing, static analysis, and code reviews. 7. Document compiler architecture, APIs, and tuning guides for internal and open‑source users. Required Skills: - Proficiency in C++ (≥ 11) and Python; experience building complex systems bridging high‑level frameworks to low‑level runtimes. - Deep understanding of compiler theory, IR design, and MLIR framework APIs. - Practical experience optimizing computational graphs and implementing custom dialects/transformation passes. - Familiarity with AI frameworks: PyTorch, TensorFlow, JAX, or similar. - Strong background in systems programming and low‑level hardware interaction (RISC‑V, ASICs, or GPUs). - Experience with open‑source project management, version control (Git), CI/CD pipelines, and unit/integration testing. - Excellent problem‑solving, communication, and teamwork skills. Required Education & Certifications: - Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, or a related technical field. - Professional certifications in compiler development, AI/ML engineering, or related domains are a plus.
Toronto, Canada
Hybrid
03-03-2026
Company background Company brand
Company Name
Tenstorrent
Job Title
Software Engineer, Metal Runtime
Job Description
Job title: Software Engineer, Metal Runtime Role Summary: Design, build, and optimize low‑level runtime systems that operate directly on AI accelerator hardware. Deliver high‑performance, parallel, bare‑metal software to manage compute units, data flow, and asynchronous execution on RISC‑V based ASICs. Expectations: - Implement fast, scalable runtime components in C/C++ for multi‑processor environments. - Collaborate closely with hardware teams to leverage architectural features. - Diagnose, profile, and tune performance across massively parallel systems. - Maintain code quality through rigorous testing and code reviews. Key Responsibilities: - Develop and refine metal‑level runtime libraries for AI accelerators. - Create bare‑metal drivers to control compute cores, memory traffic, and interrupt handling. - Interface with hardware teams to understand and exploit new architectural capabilities. - Profile and optimize thread scheduling, memory movement, and vectorization. - Debug complex, performance‑critical software across a heterogeneous chip. - Document design decisions, APIs, and tuning guidelines. Required Skills: - Proficient in C/C++ with experience in low‑level programming. - Strong understanding of multiprocessing, thread synchronization, and memory hierarchy. - Experience writing parallel, high‑throughput code for multi‑core systems. - Familiarity with RISC‑V, ASIC or FPGA programming desirable. - Ability to debug and analyze performance using tools such as perf, gdb, and hardware analyzers. - Excellent problem‑solving, communication, and teamwork skills. Required Education & Certifications: - Bachelor’s degree (or equivalent experience) in Computer Science, Electrical Engineering, or related field. - No specific certifications required; knowledge of CPU architecture and compiler toolchains preferred.
Toronto, Canada
On site
03-03-2026