AI Research Resident (Algorithms Codesign)

About the role

NORMAL COMPUTING | BUILD WITH US

Normal is an applied AI company solving the hardest problems in AI and silicon. We build foundational hardware and software for the semiconductor industry, critical AI infrastructure, and the broader systems that power our world, in partnership with the world's most advanced institutions. We work as one team across New York City, Silicon Valley (Mountain View), and London.

THE RESIDENCY PROGRAM

The AI Research Residency is Normal Computing's flagship program for exceptional researchers and engineers who want to work at the frontier of AI. Residents join a small, hand-picked cohort with a dedicated research mentor, direct access to the team developing the algorithms for our in-memory inference hardware, and a clear arc from onboarding through publication.

Every residency is built around two milestones that mark you as part of something bigger than a single project: a research paper co-authored with our team, and a company-wide research colloquium where you present your findings to the full Normal Computing organization. You'll leave with a body of published, presented work, and a standing as one of the earliest residents to help define what this program becomes.

THIS TRACK: ALGORITHMS CODESIGN

Normal's hardware runs the most memory-bound parts of large-model inference inside memory itself, rather than shuttling weights to a processor. The software that makes this programmable doesn't exist yet in any standard form. This track is about the computational methods that make LLM inference run efficiently on it: rethinking how operations like attention, memory access, and long-context decoding should behave on hardware that computes inside memory, rather than adapting standard GPU kernels to a new chip.

You'll work on this as a co-design problem. The hardware and the algorithms are developed in parallel, so your findings feed architectural decisions rather than implementing against a fixed specification. You'll validate ideas against our simulators and FPGA prototypes first, and against silicon as it arrives.

YOUR NORMAL EXPERIENCE

This is a hands-on research residency: you'll take ownership of a real technical problem in inference algorithms for in-memory compute, work alongside the researchers and engineers building our platform, and be expected to contribute ideas, not just execute someone else's. Your job is to help turn research into production-quality research code that a team can build on. Outside-of-the-box thinking is encouraged.

You'll get direct exposure to the pace, ambiguity, and speed of decision-making that comes with working at a fast-moving, well-funded startup, where the distance between an idea on a whiteboard and a result on real hardware is measured in weeks, not years.

WHAT YOU'LL DO

Develop inference algorithms for in-memory hardware. Understand which parts of LLM workloads are best suited to in-memory execution and develop algorithms that exploit it natively, demonstrated first in simulation and then on hardware.