Top Tech Jobs & Startup Jobs

Build and deploy LLM agents, agentic tools, evaluation systems, and execution environments for the kernels engineering team. Responsibilities include integrating tool-calling and MCP workflows, enabling agents to verify work using simulators, compilers, profilers, and hardware, shipping internal products and infrastructure, managing data pipelines, and measuring tool effectiveness. The role also supports founder-led projects and requires strong software engineering, product judgment, and comfort working across AI, hardware, and infrastructure domains.

Lead the technical direction of custom AI accelerator simulation infrastructure. Guide development of bit-accurate and cycle-aware models, oversee QEMU and multi-chip integration, establish validation and testing practices, and partner with hardware, firmware, kernel, and software teams on hardware-software co-design. The role includes mentoring engineers, leading design and code reviews, solving complex debugging and performance issues, and remaining hands-on with critical implementations and architectural prototypes.

Own internal rack and subassembly builds from planning through delivery. Coordinate materials, staffing, schedules, procedures, tooling, testing, and production transitions. Integrate mechanical, thermal, power, and electrical systems; review designs for manufacturability; lead technicians; resolve build issues; track risks and blockers; and improve processes for NPI and higher-volume production.

Own the full employee lifecycle, including onboarding, transfers, leaves, offboarding, benefits, payroll inputs, compliance, and HRIS administration. Advise employees and managers on policies, pay, performance, and organizational changes. Lead performance, engagement, recognition, and learning programs while partnering with Finance, Legal, brokers, and vendors. Improve People systems, automate processes, maintain accurate records, and support compensation, equity, relocation, and housing programs.

Build and deploy production-level solutions for Etched’s critical engineering challenges. Projects may include internal hardware tools, manufacturing execution systems, or agentic AI systems for R&D. The intern will own a project from scoping through production, working across full-stack engineering, infrastructure, and product development in a fast-paced, ambiguous environment.

Deploy, validate, and debug accelerator-based systems in large-scale datacenters. Diagnose issues across hardware, firmware, drivers, Linux, and networking; coordinate updates; perform root-cause analysis; build diagnostics and monitoring; create deployment documentation; and improve scalable deployment processes with customers and cross-functional engineering teams.

Lead deployment, installation, bring-up, validation, and customer acceptance of accelerator-based datacenter systems. Diagnose hardware, firmware, Linux, driver, and networking issues; perform root-cause analysis; develop deployment tooling and diagnostics; and coordinate firmware updates and hotfixes. Collaborate with engineering and sustaining teams to improve reliability and minimize downtime. The role requires up to 30% domestic and international travel.

Designs and executes automated supercomputing test suites for inference server hardware and software. Responsibilities include testing boot, system management, workloads, performance, and fault scenarios; analyzing metrics and hardware logs; debugging failures; collaborating with engineering and ODM teams; maintaining portable testing infrastructure; documenting test results; and improving test processes.

Build software-defined infrastructure for hybrid HPC clusters supporting ASIC simulation, synthesis, emulation, and CI. Design orchestration layers, infrastructure control planes, workload migration systems, and large-scale observability platforms with telemetry, tracing, alerting, synthetic testing, and fault injection. The role requires production-grade systems programming, infrastructure as code, cloud and bare-metal expertise, and the ability to debug complex hardware-software environments while mentoring engineers.

Train and post-train frontier models for engineering tasks, owning the post-training stack end to end. Build reinforcement-learning environments and verifiable reward systems with ASIC, kernel, and supercomputing engineers; develop custom data mixes, benchmarks, and specialized production models. Optimize RL systems for speed and cost while validating rollouts and performance metrics. The role is fully in person in San Jose.