Research Intern - Loop Engineering

SK hynix•San Jose, CA

1d•Onsite

About The Position

We are hiring a research intern in loop engineering: best practices for closed-loop flows that automate, optimize, and explore SoC designs. These are AI-assisted hardware/software co-design flows. CHIA, from UC Berkeley, is one example of a framework you may use. You will own a scoped research question with a full-time researcher: build or extend the loop, run the experiments, and write up what the results mean for design automation.

Requirements

Progress toward a bachelor’s, master’s, or Ph.D. in electrical engineering, computer engineering, computer science, or a related field.
Coursework or project experience in loop engineering, computer architecture, digital design, VLSI, or electronic design automation.
Fluency in Python, and comfort reading and changing an existing codebase.
Ability to pose a small research question, keep a record of what changed, and explain the result.

Nice To Haves

Exposure to RTL (Verilog, SystemVerilog, Chisel, or XLS) or to an SoC design framework.
Experience with an architecture simulator or with RTL simulation.
Familiarity with an AI-assisted hardware/software co-design loop framework, such as CHIA, or with scripted hardware design, verification, or design-space exploration flows.

Responsibilities

Study how a design loop should be structured: which steps belong in it, what feedback each step returns, and how search, verification, and physical-design results drive the next iteration.
Implement the loop with languages and tools such as Chisel and XLS, together with simulation, software build, and physical-design tools.
Apply it to an SoC problem such as microarchitectural exploration, RTL implementation and verification, or power, performance, and area optimization.
Compare loop designs on the quality of the design point, the cost of the search, reproducibility, reuse, and how reliably a failure is diagnosed.
Make the loop reliable enough to rerun: clear interfaces between steps, cached results, regression checks, and reports that support a conclusion.