PhD position in scalable ML in programmable switches

Networks Data Science Group

Deadline for receipt of applications: February 8th, 2024 23:59 AoE (09 February 2024, 13:59h Europe/Madrid Time)

The Networks Data Science group (http://nds.networks.imdea.org/) at IMDEA Networks Institute has an opening for one PhD student in the area of network intelligence. The successful candidate will design original solutions to make ML models deployed in programmable switches highly scalable. The PhD studies will build upon and extend recent research by the NDS group [1,2] with the goal to perform line-rate inference at up to 100 Gbps. The candidate will implement and evaluate the solutions in a state-of-the-art experimental platform including a variety of programmable user-plane equipment, so as to demonstrate viability and performance in industry-grade environments.

The position offers:

• hands-on training in network programming via the P4 language
• hands-on training in AI/ML design and integration in programmable user planes
• an opportunity to work with production-grade experimental platforms
• a vibrant, collaborative, multi-cultural and English-speaking environment
• the prospect to publish at top-tier venues in networking
• an advantageous path to a successful career in industry or academia [3]
• the high quality of life of the region of Madrid, Spain, where we are based.

The position requires:

• a degree in Computer Science or related field, with a solid academic record
• excellent programming skills
• a strong interest in applied AI/ML and systems (prior experience is a plus)
• fluency in written and spoken English
• enthusiasm for interdisciplinary research.

Inquiries on the position can be directed to the thesis supervisor, Dr. Marco Fiore, via email at marco.fiore@imdea.org.

Candidates shall submit by the call deadline a CV, a motivation letter, and the contact details of two references through the IMDEA Networks Institute hiring portal, at https://careers.networks.imdea.org/.

[1] A.T.-J. Akem, M. Gucciardo, M. Fiore, Flowrest: Practical Flow-Level Inference in Programmable Switches with Random Forests, IEEE INFOCOM 2023

[2] A.T.-J. Akem, M. Gucciardo, M. Fiore, Jewel: Resource-Efficient Joint Packet and Flow Level Inference in Programmable Switches IEEE INFOCOM 2024

[3] https://networks.imdea.org/team/imdea-networks-team/alumni-network/

This position will be co-funded by project PCI2022-133013 (ECOMOME: Energy Consumption Measurements and Optimization in Mobile Networks), funded by MCIN/AEI/10.13039/501100011033 and the European Union “NextGenerationEU”/PRTR.

Equal Employment Opportunity

IMDEA Networks Institute aims to increase the proportion of women and therefore qualified female applicants are explicitly encouraged to apply. Until a balanced ratio of men and women has been achieved at the institute, preference will be given to women if applicants have similar qualifications. IMDEA Networks Institute actively promotes diversity and equal opportunities. Applicants are not to be discriminated against in personnel selection procedures on the grounds of gender, ethnicity, religion or ideology, age, sexual orientation (anti-discrimination). People with disabilities who have the relevant qualifications are expressly invited to apply.