Dear all,
I am looking for a
2-year postdoc to do research on efficient foundation models at the University of Edinburgh!
As a postdoc, you will conduct blue-sky research and help us design a new benchmark for foundation models to identify optimal trade-offs between accuracy, efficiency, and cost.
This will inform the design of the next generation of hardware, streamlining accelerators for real-world AI use cases, and contribute to the development of green and energy-efficient AI models.
The postdoc will be supervised by me (Edoardo Ponti) and will become part of a wide research team that includes Luo Mai, Wenda Li, and Boris Grot (Edinburgh Informatics), Yiren (Aaron) Zhao (Imperial College London), Adrian Jackson and Nick Brown (EPCC).
The deadline for applications is the
30th of September! The position is full-time and in-person (or hybrid). The prospective start date is January 2026.
This is part of Advanced Research + Invention Agency (ARIA)’s funding for Scaling Compute: AI at 1/1000th the cost.
You can find more information about the postdoc role and the application process here:
https://elxw.fa.em3.oraclecloud.com/hcmUI/CandidateExperience/en/sites/CX_1001/job/13075
Best wishes,
Edoardo Ponti
Assistant professor in Natural Language Processing, University of Edinburgh