Jim Keller joined Tenstorrent on January 6, 2021, as president, chief technology officer and a board member. Tenstorrent’s current team page identifies him as the company’s CEO. The move brought a veteran chip architect from Intel, Tesla, AMD and P.A. Semi to a startup developing computing hardware and software for AI workloads.
When Jim Keller joined Tenstorrent—and what his role was
Tenstorrent announced Keller’s appointment on January 6, 2021. At the time, the company named him president, CTO and a member of its board. Those were his incoming roles, not his current title: Tenstorrent’s current team page identifies Keller as CEO.
The announcement framed the hire around the company’s effort to build computing hardware for machine-learning workloads. Tenstorrent’s then-CEO, Ljubisa Bajic, said the company believed “the ongoing shift towards ML-centric software necessitates a corresponding transformation in computational capabilities.” Keller described the opportunity this way: “Software 2.0 is the largest opportunity for computing innovation in a long time. Victory requires a comprehensive re-thinking of compute and low level software.” Both statements appeared in the company’s January 2021 announcement; they express the speakers’ rationale for the move.
Who is Jim Keller?
Keller is a semiconductor engineer and architect whose documented career spans processor design, system-on-chip engineering and leadership roles at several chip companies. Intel’s April 2018 announcement said he would lead silicon engineering, including system-on-chip development and integration. He joined Intel from Tesla, where he had been vice president of Autopilot and Low Voltage Hardware.
#1 Best Overall
- NVIDIA Volta GV100 Architecture — 4,608 CUDA Cores, 640 1st-Gen Tensor Cores delivering 14 TFLOPS FP32 and 112 TFLOPS deep learning performance for AI training, inference, HPC, and scientific computing workloads
- 32GB HBM2 ECC Memory — 900 GB/s Bandwidth — High-bandwidth memory on a 4096-bit bus with ECC error correction provides the memory capacity and throughput required for the largest AI models, simulations, and datasets
- PCIe 3.0 x16 Interface — 250W TDP — Standard PCIe Gen3 connectivity with passive cooling designed for enterprise rack server deployment in HPE ProLiant, Dell PowerEdge, and Supermicro platforms with adequate chassis airflow
- NVLink — Scale to 96GB Unified Memory — Connect two V100 GPUs via NVLink at 300 GB/s bi-directional bandwidth to scale GPU memory from 32GB to 96GB for larger AI training and HPC workloads
- Multi-Precision Computing — Supports FP64 (7 TFLOPS), FP32 (14 TFLOPS), FP16 (112 TFLOPS) and INT8 precision modes for flexible deployment across training, inference, and scientific simulation workloads
Tenstorrent’s biography credits Keller with work on DEC Alpha, AMD K7, K8 and K12, HyperTransport, AMD Zen, Apple A4 and A5, and Tesla’s self-driving chip. These are examples of work associated with his career, not a claim that he personally designed every component of those products.
Career at a glance
- Intel: Senior vice president of Silicon Engineering; the company said in 2018 that his remit included silicon engineering, system-on-chip development and integration.
- Tesla: Vice president of Autopilot and Low Voltage Hardware immediately before joining Intel, according to Intel’s 2018 announcement.
- AMD: Corporate vice president and chief cores architect, with work that included the Zen architecture.
- P.A. Semi and Apple: Keller was P.A. Semi’s vice president of engineering and chief architect. After Apple acquired P.A. Semi, he led efforts on Apple’s A4 and A5 custom low-power mobile chips, according to Intel’s contemporaneous account.
Intel said Keller was due to start there on April 30, 2018. That timing helps place the Tenstorrent announcement in context: he joined the startup after his Intel and Tesla roles, following earlier work at AMD and P.A. Semi/Apple.
Rank #2
- High-Performance AI Processing: The MX3 is designed to handle the most demanding AI computer vision workloads, delivering exceptional performance and efficiency.
- Flexible Integration: The MX3 can be easily integrated into your existing systems via its M.2 M-key form factor and support for Linux operating systems.
- Energy Efficient: The MX3 is designed to provide high performance while minimizing power consumption.
- Comprehensive Software Development Kit (SDK): The MX3 is supported by a comprehensive SDK that simplifies development and deployment.
- Hardware compatability: The MX3 is compatible with the PCI-SIG M.2 M-key 2280 Specification. It can be used with the Raspberry Pi 5 with a M-key 2280 HAT.
What Tenstorrent makes
Tenstorrent describes itself as a company building computers for AI. Its stated work combines computer architecture, ASIC design, advanced systems and neural-network compilers. In other words, its stated scope includes both computing hardware and software used to work with neural networks.
The company’s product page lists accelerator cards, workstations and servers. The listed starting prices were $999 for cards, $9,999 for workstations and $70,000 for servers when the page was checked on September 27, 2026. These are prices displayed by Tenstorrent, not market-wide price estimates or independent comparisons; prices and availability can change.
Recommended Free Tools
Rank #3
- ✅Powered by 26 Tera-Operations Per Second (TOPS) Hailo-8 AI Processor. 2.5W typical power consumption
- ✅Scalable, enabling simultaneous processing of multi-streams & multi-models
- ✅Enabling real-time, low latency and high-efficiency AI inferencing on the edge devices
- ✅Supports TensorFlow, TensorFlow Lite, ONNX, Keras, Pytorch frameworks
- ✅Supports Linux and Windows. Supports the temperature range of -40°C to 85°C
Historical Grayskull figures are not current product guidance
An EE Times report in 2021 gave Grayskull inference processor figures of 120 cores, 65 W and 368 TOPS of inference performance. Those are historical reported specifications for that processor. They should not be treated as specifications for Tenstorrent’s current products or as an independently established comparison with other systems.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What the 2021 hiring story says—and does not say—about Tenstorrent today
The appointment showed that Tenstorrent was recruiting an experienced chip executive for its AI-computing effort. It does not, by itself, demonstrate how the company’s products perform against competitors. The company’s newsroom lists later activity, including 2026 announcements about a JapanFold drug-discovery launch with ai&, a runtime-observability partnership with Stealthium, TT-Ascalon S product and performance news, and expansion across Japan. Those announcements show that its business has continued beyond the original hiring story; they are not independent verification of technical claims.
Rank #4
- 48GB AI graphics accelerator
For a buyer or developer evaluating the company’s offerings, the relevant question is the fit of a specific current system with the intended workload, software stack, memory needs, power and cooling constraints, form factor, support and total system cost. The hiring announcement and Keller’s career history are context, not a substitute for product-specific specifications or comparative testing.
Quick Recap
Best Value
- Professional AI & Creator Workstation: AMD Radeon AI PRO R9700 GPU with 32GB GDDR6 is engineered for AI development, professional content creation, and compute-intensive workloads.
- Massive 32GB Memory Capacity: 32GB of GDDR6 memory on a 256-bit bus provides ample bandwidth for large AI models, 8K video editing, and complex 3D rendering.
- Advanced RDNA 4 with AI Accelerators: 64 Compute Units with 3rd Gen Ray Tracing and dedicated 2nd Gen AI Accelerators for groundbreaking AI performance and visual computing.
- Professional Blower Cooling: Efficient single blower design exhausts heat directly out of the chassis, ideal for multi-GPU workstation and server configurations.
- Enterprise-Grade Thermal Solution: Vapor chamber heatsink with industrial Honeywell PTM7950 thermal interface material ensures reliable cooling under sustained professional loads.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
