ASIC Architect

Cerebras Systems builds the world's largest AI chip, 56 times larger than GPUs. This architecture allows Cerebras to deliver industry-leading training and inference speeds; over 10 times faster than GPU-based hyperscale cloud inference services. This order of magnitude increase in speed is transforming the user experience of AI applications, unlocking real-time iteration and increasing intelligence via additional agentic computation. Cerebras works with the leading model labs, global enterprises, and cutting-edge AI-native startups. OpenAI recently announced a multi-year partnership https://openai.com/index/cerebras-partnership/ with Cerebras, to deploy 750 megawatts of scale, transforming key workloads with ultra high-speed inference. Responsibilities - Translate high level architecture spec to micro-architecture feature requirements - Bring up new features in the performance/power model - Perform comprehensive PPA trade-offs for new architectural features - Extract insights for new features and micro-architecture power efficiency - Profile workloads, identify bottlenecks and project competition performance for benchmarking - Engage with SW teams for end-end application level modeling at cluster level - Identify kernel level HW acceleration level opportunities Qualifications - Masters/PhD in Electrical/Computer Engineering - 10+ years of experience across performance analysis and modeling across GPUs, CPUs or accelerator products - Strong background in computer architecture and key high level architectural trade-offs - Comfortable standing up new performance models from scratch in Python or similar analytical environments - Exposure to micro-code (kernel) performance bottlenecks and optimization techniques - Good understanding of how high-level workloads map to underlying micro-architecture is desired - Understanding of basic ML workload profiling techniques and model network architecture is preferred Why Join Cerebras People who are serious about software make the

Department: Hardware Departments; Team: Silicon