Note: The job is a remote job and is open to candidates in USA. Unconventional is rethinking computing from the ground up to build a more energy-efficient foundation for AI by mapping neural networks directly to semiconductor device physics. The Member of Technical Staff, System Software (Runtime & Firmware) will build runtime software, embedded firmware, drivers, and host interfaces for novel compute systems, while leading silicon bring-up and collaborating across AI, modeling, and hardware teams.
Responsibilities
- **Runtime:** Build the runtime that executes compiled models on the machine, covering dispatch, data movement, and result readback. One runtime targets both our simulated system model and real silicon, so a program runs on either unchanged
- **Embedded Firmware:** Design and implement bare-metal firmware on the on-device embedded processor, covering model execution and scheduling, device sequencing, and calibration
- **Drivers & Host Interface:** Build and maintain the drivers on both sides of the host/device boundary, along with the protocol between them, including the host-side hardware abstraction layer and the compatibility discipline required when firmware and host software ship independently
- **Silicon & Software Bring-up:** Lead post-silicon software bring-up alongside the Hardware team, own the software that drives our test chips, and close the loop between measured hardware behavior and our simulation models
- **Cross-Functional Collaboration:** Collaborate with the AI, System Modeling, and Hardware teams, our first customers, to get prototype hardware into their hands and keep it running as the platform scales
Skills
- B.S. or M.S. in Electrical Engineering, Computer Engineering, Computer Science, or a related field
- 7+ years of industry experience in embedded firmware, driver development, or hardware-adjacent systems software, including at least one bring-up of new or prototype silicon and hardware that you took from power-on to running real workloads
- Expertise in C/C++ on embedded processors, including interrupts, DMA, memory-mapped register access, and boot and flash paths
- Experience developing Linux drivers in kernel or user space, with demonstrated depth in at least one of PCIe, USB, UART, or SPI, and in designing robust protocols over an imperfect physical link
- Strong Python engineering alongside your C/C++, and the versioning discipline required when firmware, drivers, and host software each ship independently
- Ability to identify, triage, and solve anomalies across silicon, board, firmware, and software, and to correlate measured behavior against simulation models
- Excellent ability to communicate complex technical concepts to diverse teams across the full stack from algorithms to circuits
- Experience building the runtime or driver layer for an AI accelerator, NPU, or GPU, including kernel dispatch, queueing, and memory management
- Experience designing instruction encodings, microcode, or control-flow sequencing for a non-CPU compute engine
- Hands-on experience developing, integrating, and debugging RTL on FPGA platforms, including timing closure and bring-up of prototype boards, plus familiarity with embedded microcontroller IP, PCIe, and the host driver stacks that target such platforms
- Familiarity with cloud infrastructure and deployment, including virtualization, containerization, and running hardware-attached services
- Any exposure to analog, mixed-signal, and/or other unconventional physics-based computing paradigms
Benefits
- Best-in-class health benefits
- 401k matching
- Truly unlimited PTO
- Complimentary meals in our Mountain View office
Company Overview