AI Research Resident (Algorithms Codesign)

What you'll do

• Develop inference algorithms and numerical methods for in-memory hardware, validating them in simulation and on hardware.
• Build benchmarks and evaluation frameworks, and translate workload findings into hardware and architecture constraints.
• Lead an original research investigation and co-author a paper; present findings at the residency colloquium.

What they ask for

• Currently pursuing or recently completed a graduate degree (MS or PhD, or equivalent research experience) in a related field.
• Working understanding of large-model inference, including attention, KV cache, long-context decoding, and memory bandwidth constraints.
• Strong Python skills and some experience with a systems language or implementing algorithms close to hardware.
• Experience turning a research idea into working, reasonably reproducible code.

Nice to have

• Exposure to inference optimization, stochastic or probabilistic methods, numerical analysis, or analog computation.
• Exposure to analog or mixed-signal systems, in-memory compute, or non-von-Neumann architectures; publications or open-source work in relevant areas.

Pay

$150K (currency not stated; pay type not stated)

Tools

Python, C++, Rust

AI focus

Research and develop efficient LLM inference algorithms for in-memory hardware, including methods that exploit thermal noise and analog dynamics.