Hybrid P4 Offload Infra Intern: Low-Level Networking

Kalray; Montbonnot-Saint-Martin; Remote; EUR 8,900 - 13,000; Full time

Benefits offered by this job

Hybrid work

Meal vouchers

Team building activities

Job summary

Kalray invites an engineering student for a 6-month internship to help develop a P4 offload infrastructure on a high-performance network card. You will work across low-level C, embedded firmware, and Linux kernel interfaces to translate P4 descriptions into card-executable processes.

The project spans the toolchain, embedded execution, and host configuration, with hands-on experience in compilation, hardware interactions, and performance measurements in throughput and latency.

Qualifications

• C language proficiency and strong debugging skills.
• Basic knowledge of C++ and interest in toolchains.
• Familiarity with processor architecture and memory management.
• Experience with Linux debugging and software/hardware interface concepts.
• Comfortable using English in technical contexts.

Responsibilities

• Contribute to defining the P4 execution model on the target and mapping program elements to hardware resources.
• Participate in defining P4 features and hardware constraints for the target.
• Help develop a compilation backend to translate IR to code for the custom processor.
• Assist in implementing table/action lookup, memory resource management, and host-card interfaces.
• Support firmware integration and end-to-end tests, including performance measurements.

Skills

C programming

Low-level programming

C++ basics

Linux debugging

English in tech context

Education

Engineering student

Job description

Kalray invites an engineering student for a 6-month internship to help develop a P4 offload infrastructure on a high-performance network card. You will work across low-level C, embedded firmware, and Linux kernel interfaces to translate P4 descriptions into card-executable processes.

The project spans the toolchain, embedded execution, and host configuration, with hands-on experience in compilation, hardware interactions, and performance measurements in throughput and latency.