AI Research Resident (Algorithms Codesign)
Originally posted October 27, 2026 by the employer.
About the role
As an AI Research Resident in Algorithms Codesign at Normal Computing, you will develop computational methods and inference algorithms tailored for in-memory compute hardware. You will investigate how operations like attention, memory access, and long-context decoding run on non-von-Neumann hardware, validating designs across simulators, FPGA prototypes, and silicon.
What you'll do
- Develop inference algorithms for in-memory hardware that address memory-bound parts of LLM workloads, validating results first in simulation and then on physical hardware.
- Design numerical methods that exploit thermal noise and analog dynamics rather than treating them as constraints to work around.
- Build evaluation suites, frameworks, and benchmarks to characterize algorithmic behavior across simulators, FPGA prototypes, and silicon.
- Translate large model workload insights into architectural constraints and design opportunities for the hardware team.
- Execute an original research investigation, co-authoring a research paper for venue submission and presenting findings at the residency research colloquium.
What you'll need
- Graduate standing (currently pursuing or recently completed MS or PhD, or equivalent research experience) in machine learning, applied mathematics, physics, electrical engineering, computer science, or a related field.
- Working understanding of large model inference concepts, including attention mechanisms, KV cache, long-context decoding, and memory bandwidth constraints.
- Exposure to inference optimization techniques (quantization, sparsity, kernel fusion, or memory-efficient attention) or to stochastic systems, probabilistic methods, numerical analysis, or analog computation.
- Python programming skills combined with experience in a systems language (C++, Rust, or similar) or implementing algorithms close to hardware.
- Demonstrated experience translating research concepts into reproducible code.
Nice to have
- Exposure to analog or mixed-signal systems, in-memory compute, or non-von-Neumann architectures.
- Publications or open-source contributions in efficient inference, stochastic algorithms, or novel computing substrates.
- Prior experience working on hardware platforms that did not yet exist when you joined.
Skills: in-memory computing, LLM inference, hardware-software codesign, quantization, FPGA
More semiconductor jobs: Normal Computing jobs · AI chip design jobs
Tracked in our index since 28 Oct 2026. This is our observation date, not the employer posting date. Specialty: AI/ML Hardware Engineering (AI ML Hardware). 483 active roles in this specialty at 76 employers in our index, as of 20 Nov 2026.
How we counted: postings read from employer career sites. How we count