Inference Intern
Etched · San Jose
About this role
## Etched — Inference Architecture Intern (Spring ’27 / Summer ’27) ### About Etched Etched is building hardware for frontier intelligence—co-designing chips, racks, software, and manufacturing to deliver best-in-class throughput and latency across inference workloads. Backed by top-tier investors and staffed by leading engineers, Etched is redefining the infrastructure layer for the fastest-growing industry in history. ### Job Summary Etched is seeking **Spring ’27 and Summer ’27 Inference Architecture interns** to help design and optimize **next-generation AI accelerator compute architectures** for exceptional performance and efficiency on inference workloads. You’ll work on cutting-edge architectural problems and performance modeling throughout your internship. ### Key Responsibilities - Support porting state-of-the-art models to Etched’s architecture - Build programming abstractions and testing capabilities to rapidly iterate on model porting - Build, enhance, and scale the runtime (multi-node inference, intra-node execution, state management, robust error handling) - Optimize routing and communication layers using collectives - Use performance profiling and debugging tools to identify bottlenecks and correctness issues - Co-design HW instructions and model architecture operations to maximize model performance - Implement high-performance software components for the Model Toolkit ### You May Be a Good Fit If You Have - Progress toward a **Bachelor’s, Master’s, or PhD** in CS, computer engineering, applied math, or related field - Proficiency in **Python** and **C++** - Understanding of performance-sensitive/distributed systems (e.g., Linux internals, accelerator architectures like GPUs/TPUs, compilers, high-speed interconnects like NVLink/InfiniBand) - Experience porting applications to non-standard accelerator hardware/platforms - Deep knowledge of **transformer architectures** and/or inference serving stacks (e.g., **vLLM, SGLang**) ### Strong Candidates May Have - Proficiency in **Rust** - Low-latency, high-performance applications using kernel-level and user-space networking stacks - Distributed systems knowledge (consensus protocols, consistency models, communication patterns) - Solid grasp of Transformer architectures, especially **Mixture-of-Experts (MoE)** - Experience with extensive **SIMD** optimizations for performance-critical paths - Familiarity with **PyTorch** or **JAX** - Math competitions (AIME, AMC, etc.) > We encourage you to apply even if you don’t meet every qualification. ### Program Details - **12-week paid internship** - **Generous housing support** for relocating - **Daily lunch and dinner** in the office - **In-person** role based in **San Jose, CA (Santana Row)** - Direct mentorship from industry leaders and world-class engineers - Opportunity to work on one of the most important problems of our time ### Contact For questions, email: **internships@etched.com**
Listing freshness
CronJobs last confirmed this listing 1h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.