Etched IA Research Engineer (Kernels) Intern role to build agents that accelerate frontier token production. Work embedded with kernels team to develop tools, execution environments, and internal products for engineering workflows.
Responsibilities
Embed with our kernels team to find high-leverage bottlenecks and help turn them into tools engineers use every day.
Build and deploy LLM agents wired into our engineering workflows via tool-calling and MCP.
Help build execution environments where agents can verify their own work, with sandboxed access to simulators, compilers, profilers, and real hardware.
Ship internal products end to end, including the interfaces, data pipelines, and infrastructure behind them.
Design evals and track adoption to measure which tools are working and which aren’t.
Support special projects with the founders across the company.
You may be a good fit if you have
Progress towards a Bachelor’s, Master’s, or PhD degree in computer science, computer engineering, or a related field
Strong software engineering fundamentals and a record of shipping projects other people use
Built agents for multi-step tasks beyond calling an API: tool design, context management, evals, and failure analysis
Heavy use of coding agents, including reading their transcripts and changing the system when an agent is confidently wrong
High agency in unfamiliar domains, product sense, and extreme attention to detail
Qualification
Progress towards Bachelor’sHigh agency in unfamiliar domainsChip developmentInfrastructureFine-tuningHackathons
Required
Progress towards Bachelor’s, Master’s, or PhD in computer science, computer engineering, or related field
Strong software engineering fundamentals and record of shipping projects
Experience building agents for multi-step tasks beyond API calls
Experience with coding agents including transcript analysis and system changes
High agency in unfamiliar domains, product sense, and attention to detail
Preferred
Chip development, GPU kernels, hardware, or manufacturing experience
Infrastructure, data engineering, or DevOps experience
Fine-tuning, RAG over technical data, or multi-agent orchestration
Hackathons, startups, or side projects with real users
