Chip Simulation Software Engineer

Etched is building the world’s first AI inference system purpose-built for transformers - delivering over 10x higher performance and dramatically lower cost and latency than a B200. With Etched ASICs, you can build products that would be impossible with GPUs, like real-time video generation models and extremely deep & parallel chain-of-thought reasoning agents. Backed by hundreds of millions from top-tier investors and staffed by leading engineers, Etched is redefining the infrastructure layer for the fastest growing industry in history.

We are seeking highly motivated and detail-oriented software engineers to join our chip simulation team. We’re building Sohu, the world’s first transformer ASIC that leverages tight hardware-software co-design to deliver world-best performance. The chip simulation team plays a key role in hardware-software co-design, enabling early development of the full Sohu software stack. As a software engineer on this team, you will design, develop, and test simulations of our custom hardware systems. You will be working at the cutting edge of hardware innovation, ensuring that our software and hardware teams are able to closely coordinate on their development paths.

• →

• Proficiency in C/C++

• Strong understanding of low-level software engineering.

• Strong understanding of hardware engineering.

• Excellent communication and collaboration skills.

• Experience developing hardware or system simulations (e.g., SystemC, gem5, QEMU, or custom frameworks).

• Experience with custom ML hardware accelerators (e.g., Google TPU)

• Experience with firmware, kernel and driver development (e.g., Linux, FreeRTOS)

• Experience with hardware development and verification.

• Experience with multiprocessing and multithreading.

• Familiarity with bit-accurate numerics modeling.

• Experience with performance profiling and optimization of simulation code

• Proficiency with Python for test harness development, bazel, and git.

• Implement a simulation of Sohu’s on-chip modules.

• Integrate simulated chips with simulated hosts (QEMU), enabling Kernel driver development and host software testing.

• Integrate multiple simulated chips to validate multi-chip software scale-out and scale-up approaches.

• Work closely with Sohu software teams in debugging early software deliverables.

Etched believes in the Bitter Lesson. We think most of the progress in the AI field has come from using more FLOPs to train and run models, and the best way to get more FLOPs is to build model-specific hardware. Larger and larger training runs encourage companies to consolidate around fewer model architectures, which creates a market for single-model ASICs.

We are a fully in-person team in San Jose (Santana Row), and greatly value engineering skills. We do not have boundaries between engineering and research, and we expect all of our technical staff to contribute to both as needed.