AI-Powered NoC Power Modeling Intern

QUALCOMM, Inc.; Guyancourt; On-site; EUR 10,000 - 15,000; Full time

Job summary

Qualcomm France S.A.R.L. is offering a 6-month internship in the Interns Group focused on power modeling for Network-on-Chip architectures.

You will help build a two-phase framework combining traditional power characterization with machine-learning based modeling to enable fast design-space exploration. The project covers data preparation, including NoC IP architectures, traffic scenarios, and power characterization through synthesis and simulations.

Qualifications

• Final year master’s degree in Microelectronics, Computer Engineering or related field.
• Proficient in Python and Bash scripting; knowledge of Verilog/VHDL, UVM, EDA flows and ML frameworks is a plus.
• Proficient in English.

Responsibilities

• Phase 1: Data preparation — elaborate NoC IP architectures/topologies and define representative traffic.
• Phase 1: Power characterization — run synthesis and simulations to capture activity profiles and power.
• Phase 1: Dataset construction — build a labelled dataset from the power characterization step.
• Phase 2: AI power modeling — train and compare AI-based models and assess accuracy against gate-level references.
• Phase 2: Early design-space exploration — use trained models to explore design in terms of power quickly.

Skills

Python scripting

Bash scripting

Verilog/VHDL

ML frameworks

English

Education

Master's degree in Microelectronics, Computer Engineering or related field

Tools

UVM

Job description

Qualcomm France S.A.R.L. is offering a 6-month internship in the Interns Group focused on power modeling for Network-on-Chip architectures.

You will help build a two-phase framework combining traditional power characterization with machine-learning based modeling to enable fast design-space exploration. The project covers data preparation, including NoC IP architectures, traffic scenarios, and power characterization through synthesis and simulations.